Add files using upload-large-folder tool

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

LICENSE

@@ -0,0 +1,49 @@
+**LLAMA 3.3 COMMUNITY LICENSE AGREEMENT**
+
+ Llama 3.3 Version Release Date: December 6, 2024
+
+“**Agreement**” means the terms and conditions for use, reproduction, distribution and modification of the Llama Materials set forth herein.
+
+“**Documentation**” means the specifications, manuals and documentation accompanying Llama 3.3 distributed by Meta at [https://www.llama.com/docs/overview](https://llama.com/docs/overview).
+
+“**Licensee**” or “**you**” means you, or your employer or any other person or entity (if you are entering into this Agreement on such person or entity’s behalf), of the age required under applicable laws, rules or regulations to provide legal consent and that has legal authority to bind your employer or such other person or entity if you are entering in this Agreement on their behalf.
+
+“**Llama 3.3**” means the foundational large language models and software and algorithms, including machine-learning model code, trained model weights, inference-enabling code, training-enabling code, fine-tuning enabling code and other elements of the foregoing distributed by Meta at [https://www.llama.com/llama-downloads](https://www.llama.com/llama-downloads).
+
+“**Llama Materials**” means, collectively, Meta’s proprietary Llama 3.3 and Documentation (and any portion thereof) made available under this Agreement.
+
+“**Meta**” or “**we**” means Meta Platforms Ireland Limited (if you are located in or, if you are an entity, your principal place of business is in the EEA or Switzerland) and Meta Platforms, Inc. (if you are located outside of the EEA or Switzerland). 
+
+By clicking “I Accept” below or by using or distributing any portion or element of the Llama Materials, you agree to be bound by this Agreement.
+
+1\. **License Rights and Redistribution**.
+
+a. Grant of Rights. You are granted a non-exclusive, worldwide, non-transferable and royalty-free limited license under Meta’s intellectual property or other rights owned by Meta embodied in the Llama Materials to use, reproduce, distribute, copy, create derivative works of, and make modifications to the Llama Materials.  
+
+b. Redistribution and Use.  
+
+i. If you distribute or make available the Llama Materials (or any derivative works thereof), or a product or service (including another AI model) that contains any of them, you shall (A) provide a copy of this Agreement with any such Llama Materials; and (B) prominently display “Built with Llama” on a related website, user interface, blogpost, about page, or product documentation. If you use the Llama Materials or any outputs or results of the Llama Materials to create, train, fine tune, or otherwise improve an AI model, which is distributed or made available, you shall also include “Llama” at the beginning of any such AI model name.
+
+ii. If you receive Llama Materials, or any derivative works thereof, from a Licensee as part of an integrated end user product, then Section 2 of this Agreement will not apply to you. 
+
+iii. You must retain in all copies of the Llama Materials that you distribute the following attribution notice within a “Notice” text file distributed as a part of such copies: “Llama 3.3 is licensed under the Llama 3.3 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.”
+
+iv. Your use of the Llama Materials must comply with applicable laws and regulations (including trade compliance laws and regulations) and adhere to the Acceptable Use Policy for the Llama Materials (available at [https://www.llama.com/llama3\_3/use-policy](https://www.llama.com/llama3_3/use-policy)), which is hereby incorporated by reference into this Agreement.  
+    
+2\. **Additional Commercial Terms**. If, on the Llama 3.3 version release date, the monthly active users of the products or services made available by or for Licensee, or Licensee’s affiliates, is greater than 700 million monthly active users in the preceding calendar month, you must request a license from Meta, which Meta may grant to you in its sole discretion, and you are not authorized to exercise any of the rights under this Agreement unless or until Meta otherwise expressly grants you such rights.
+
+3**. Disclaimer of Warranty**. UNLESS REQUIRED BY APPLICABLE LAW, THE LLAMA MATERIALS AND ANY OUTPUT AND RESULTS THEREFROM ARE PROVIDED ON AN “AS IS” BASIS, WITHOUT WARRANTIES OF ANY KIND, AND META DISCLAIMS ALL WARRANTIES OF ANY KIND, BOTH EXPRESS AND IMPLIED, INCLUDING, WITHOUT LIMITATION, ANY WARRANTIES OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE. YOU ARE SOLELY RESPONSIBLE FOR DETERMINING THE APPROPRIATENESS OF USING OR REDISTRIBUTING THE LLAMA MATERIALS AND ASSUME ANY RISKS ASSOCIATED WITH YOUR USE OF THE LLAMA MATERIALS AND ANY OUTPUT AND RESULTS.
+
+4\. **Limitation of Liability**. IN NO EVENT WILL META OR ITS AFFILIATES BE LIABLE UNDER ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, TORT, NEGLIGENCE, PRODUCTS LIABILITY, OR OTHERWISE, ARISING OUT OF THIS AGREEMENT, FOR ANY LOST PROFITS OR ANY INDIRECT, SPECIAL, CONSEQUENTIAL, INCIDENTAL, EXEMPLARY OR PUNITIVE DAMAGES, EVEN IF META OR ITS AFFILIATES HAVE BEEN ADVISED OF THE POSSIBILITY OF ANY OF THE FOREGOING.
+
+5\. **Intellectual Property**.
+
+a. No trademark licenses are granted under this Agreement, and in connection with the Llama Materials, neither Meta nor Licensee may use any name or mark owned by or associated with the other or any of its affiliates, except as required for reasonable and customary use in describing and redistributing the Llama Materials or as set forth in this Section 5(a). Meta hereby grants you a license to use “Llama” (the “Mark”) solely as required to comply with the last sentence of Section 1.b.i. You will comply with Meta’s brand guidelines (currently accessible at [https://about.meta.com/brand/resources/meta/company-brand/](https://about.meta.com/brand/resources/meta/company-brand/)[)](https://en.facebookbrand.com/). All goodwill arising out of your use of the Mark will inure to the benefit of Meta.
+
+b. Subject to Meta’s ownership of Llama Materials and derivatives made by or for Meta, with respect to any derivative works and modifications of the Llama Materials that are made by you, as between you and Meta, you are and will be the owner of such derivative works and modifications.
+
+c. If you institute litigation or other proceedings against Meta or any entity (including a cross-claim or counterclaim in a lawsuit) alleging that the Llama Materials or Llama 3.3 outputs or results, or any portion of any of the foregoing, constitutes infringement of intellectual property or other rights owned or licensable by you, then any licenses granted to you under this Agreement shall terminate as of the date such litigation or claim is filed or instituted. You will indemnify and hold harmless Meta from and against any claim by any third party arising out of or related to your use or distribution of the Llama Materials.
+
+6\. **Term and Termination**. The term of this Agreement will commence upon your acceptance of this Agreement or access to the Llama Materials and will continue in full force and effect until terminated in accordance with the terms and conditions herein. Meta may terminate this Agreement if you are in breach of any term or condition of this Agreement. Upon termination of this Agreement, you shall delete and cease use of the Llama Materials. Sections 3, 4 and 7 shall survive the termination of this Agreement. 
+
+7\. **Governing Law and Jurisdiction**. This Agreement will be governed and construed under the laws of the State of California without regard to choice of law principles, and the UN Convention on Contracts for the International Sale of Goods does not apply to this Agreement. The courts of California shall have exclusive jurisdiction of any dispute arising out of this Agreement.   

README.md

@@ -0,0 +1,312 @@
+---
+library_name: transformers
+language:
+- en
+- fr
+- it
+- pt
+- hi
+- es
+- th
+- de
+base_model:
+- meta-llama/Llama-3.1-70B
+tags:
+- facebook
+- meta
+- pytorch
+- llama
+- llama-3
+extra_gated_prompt: "### LLAMA 3.3 COMMUNITY LICENSE AGREEMENT\nLlama 3.3 Version Release Date: December 6, 2024\n\"Agreement\" means the terms and conditions for use, reproduction, distribution and modification of the Llama Materials set forth herein.\n\"Documentation\" means the specifications, manuals and documentation accompanying Llama 3.3 distributed by Meta at [https://www.llama.com/docs/overview](https://llama.com/docs/overview).\n\"Licensee\" or \"you\" means you, or your employer or any other person or entity (if you are entering into this Agreement on such person or entity’s behalf), of the age required under applicable laws, rules or regulations to provide legal consent and that has legal authority to bind your employer or such other person or entity if you are entering in this Agreement on their behalf.\n\"Llama 3.3\" means the foundational large language models and software and algorithms, including machine-learning model code, trained model weights, inference-enabling code, training-enabling code, fine-tuning enabling code and other elements of the foregoing distributed by Meta at [https://www.llama.com/llama-downloads](https://www.llama.com/llama-downloads).\n\"Llama Materials\" means, collectively, Meta’s proprietary Llama 3.3 and Documentation (and any portion thereof) made available under this Agreement.\n\"Meta\" or \"we\" means Meta Platforms Ireland Limited (if you are located in or, if you are an entity, your principal place of business is in the EEA or Switzerland) and Meta Platforms, Inc. (if you are located outside of the EEA or Switzerland).\nBy clicking “I Accept” below or by using or distributing any portion or element of the Llama Materials, you agree to be bound by this Agreement.\n1. License Rights and Redistribution.\na. Grant of Rights. You are granted a non-exclusive, worldwide, non-transferable and royalty-free limited license under Meta’s intellectual property or other rights owned by Meta embodied in the Llama Materials to use, reproduce, distribute, copy, create derivative works of, and make modifications to the Llama Materials.\nb. Redistribution and Use.\ni. If you distribute or make available the Llama Materials (or any derivative works thereof), or a product or service (including another AI model) that contains any of them, you shall (A) provide a copy of this Agreement with any such Llama Materials; and (B) prominently display “Built with Llama” on a related website, user interface, blogpost, about page, or product documentation. If you use the Llama Materials or any outputs or results of the Llama Materials to create, train, fine tune, or otherwise improve an AI model, which is distributed or made available, you shall also include “Llama” at the beginning of any such AI model name.\nii. If you receive Llama Materials, or any derivative works thereof, from a Licensee as part of an integrated end user product, then Section 2 of this Agreement will not apply to you.\_\niii. You must retain in all copies of the Llama Materials that you distribute the following attribution notice within a “Notice” text file distributed as a part of such copies: “Llama 3.3 is licensed under the Llama 3.3 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.”\niv. Your use of the Llama Materials must comply with applicable laws and regulations (including trade compliance laws and regulations) and adhere to the Acceptable Use Policy for the Llama Materials (available at [https://www.llama.com/llama3\\_3/use-policy](https://www.llama.com/llama3_3/use-policy)), which is hereby incorporated by reference into this Agreement.  \n2. Additional Commercial Terms. If, on the Llama 3.3 version release date, the monthly active users of the products or services made available by or for Licensee, or Licensee’s affiliates, is greater than 700 million monthly active users in the preceding calendar month, you must request a license from Meta, which Meta may grant to you in its sole discretion, and you are not authorized to exercise any of the rights under this Agreement unless or until Meta otherwise expressly grants you such rights.\n3. Disclaimer of Warranty. UNLESS REQUIRED BY APPLICABLE LAW, THE LLAMA MATERIALS AND ANY OUTPUT AND RESULTS THEREFROM ARE PROVIDED ON AN “AS IS” BASIS, WITHOUT WARRANTIES OF ANY KIND, AND META DISCLAIMS ALL WARRANTIES OF ANY KIND, BOTH EXPRESS AND IMPLIED, INCLUDING, WITHOUT LIMITATION, ANY WARRANTIES OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE. YOU ARE SOLELY RESPONSIBLE FOR DETERMINING THE APPROPRIATENESS OF USING OR REDISTRIBUTING THE LLAMA MATERIALS AND ASSUME ANY RISKS ASSOCIATED WITH YOUR USE OF THE LLAMA MATERIALS AND ANY OUTPUT AND RESULTS.\n4. Limitation of Liability. IN NO EVENT WILL META OR ITS AFFILIATES BE LIABLE UNDER ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, TORT, NEGLIGENCE, PRODUCTS LIABILITY, OR OTHERWISE, ARISING OUT OF THIS AGREEMENT, FOR ANY LOST PROFITS OR ANY INDIRECT, SPECIAL, CONSEQUENTIAL, INCIDENTAL, EXEMPLARY OR PUNITIVE DAMAGES, EVEN IF META OR ITS AFFILIATES HAVE BEEN ADVISED OF THE POSSIBILITY OF ANY OF THE FOREGOING.\n5. Intellectual Property.\na. No trademark licenses are granted under this Agreement, and in connection with the Llama Materials, neither Meta nor Licensee may use any name or mark owned by or associated with the other or any of its affiliates, except as required for reasonable and customary use in describing and redistributing the Llama Materials or as set forth in this Section 5(a). Meta hereby grants you a license to use “Llama” (the “Mark”) solely as required to comply with the last sentence of Section 1.b.i. You will comply with Meta’s brand guidelines (currently accessible at [https://about.meta.com/brand/resources/meta/company-brand/](https://about.meta.com/brand/resources/meta/company-brand/)[)](https://en.facebookbrand.com/). All goodwill arising out of your use of the Mark will inure to the benefit of Meta.\nb. Subject to Meta’s ownership of Llama Materials and derivatives made by or for Meta, with respect to any derivative works and modifications of the Llama Materials that are made by you, as between you and Meta, you are and will be the owner of such derivative works and modifications.\nc. If you institute litigation or other proceedings against Meta or any entity (including a cross-claim or counterclaim in a lawsuit) alleging that the Llama Materials or Llama 3.3 outputs or results, or any portion of any of the foregoing, constitutes infringement of intellectual property or other rights owned or licensable by you, then any licenses granted to you under this Agreement shall terminate as of the date such litigation or claim is filed or instituted. You will indemnify and hold harmless Meta from and against any claim by any third party arising out of or related to your use or distribution of the Llama Materials.\n6. Term and Termination. The term of this Agreement will commence upon your acceptance of this Agreement or access to the Llama Materials and will continue in full force and effect until terminated in accordance with the terms and conditions herein. Meta may terminate this Agreement if you are in breach of any term or condition of this Agreement. Upon termination of this Agreement, you shall delete and cease use of the Llama Materials. Sections 3, 4 and 7 shall survive the termination of this Agreement.\n7. Governing Law and Jurisdiction. This Agreement will be governed and construed under the laws of the State of California without regard to choice of law principles, and the UN Convention on Contracts for the International Sale of Goods does not apply to this Agreement. The courts of California shall have exclusive jurisdiction of any dispute arising out of this Agreement.\n### Llama 3.3 Acceptable Use Policy\nMeta is committed to promoting safe and fair use of its tools and features, including Llama 3.3. If you access or use Llama 3.3, you agree to this Acceptable Use Policy (“**Policy**”). The most recent copy of this policy can be found at [https://www.llama.com/llama3\\_3/use-policy](https://www.llama.com/llama3_3/use-policy).\nProhibited Uses\nWe want everyone to use Llama 3.3 safely and responsibly. You agree you will not use, or allow others to use, Llama 3.3 to:\n1. Violate the law or others’ rights, including to:\n\n   1. Engage in, promote, generate, contribute to, encourage, plan, incite, or further illegal or unlawful activity or content, such as:  \n      1. Violence or terrorism  \n      2. Exploitation or harm to children, including the solicitation, creation, acquisition, or dissemination of child exploitative content or failure to report Child Sexual Abuse Material  \n      3. Human trafficking, exploitation, and sexual violence  \n      4. The illegal distribution of information or materials to minors, including obscene materials, or failure to employ legally required age-gating in connection with such information or materials.  \n      5. Sexual solicitation  \n      6. Any other criminal activity\n\n   2. Engage in, promote, incite, or facilitate the harassment, abuse, threatening, or bullying of individuals or groups of individuals\n\n   3. Engage in, promote, incite, or facilitate discrimination or other unlawful or harmful conduct in the provision of employment, employment benefits, credit, housing, other economic benefits, or other essential goods and services\n\n   4. Engage in the unauthorized or unlicensed practice of any profession including, but not limited to, financial, legal, medical/health, or related professional practices\n\n   5. Collect, process, disclose, generate, or infer private or sensitive information about individuals, including information about individuals’ identity, health, or demographic information, unless you have obtained the right to do so in accordance with applicable law\n\n   6. Engage in or facilitate any action or generate any content that infringes, misappropriates, or otherwise violates any third-party rights, including the outputs or results of any products or services using the Llama Materials\n\n   7. Create, generate, or facilitate the creation of malicious code, malware, computer viruses or do anything else that could disable, overburden, interfere with or impair the proper working, integrity, operation or appearance of a website or computer system\n\n   8. Engage in any action, or facilitate any action, to intentionally circumvent or remove usage restrictions or other safety measures, or to enable functionality disabled by Meta\n\n2. Engage in, promote, incite, facilitate, or assist in the planning or development of activities that present a risk of death or bodily harm to individuals, including use of Llama 3.3 related to the following:\n\n   1. Military, warfare, nuclear industries or applications, espionage, use for materials or activities that are subject to the International Traffic Arms Regulations (ITAR) maintained by the United States Department of State or to the U.S. Biological Weapons Anti-Terrorism Act of 1989 or the Chemical Weapons Convention Implementation Act of 1997\n\n   2. Guns and illegal weapons (including weapon development)\n\n   3. Illegal drugs and regulated/controlled substances\n\n   4. Operation of critical infrastructure, transportation technologies, or heavy machinery\n\n   5. Self-harm or harm to others, including suicide, cutting, and eating disorders\n\n   6. Any content intended to incite or promote violence, abuse, or any infliction of bodily harm to an individual\n\n3. Intentionally deceive or mislead others, including use of Llama 3.3 related to the following:\n\n   1. Generating, promoting, or furthering fraud or the creation or promotion of disinformation\n\n   2. Generating, promoting, or furthering defamatory content, including the creation of defamatory statements, images, or other content\n\n   3. Generating, promoting, or further distributing spam\n\n   4. Impersonating another individual without consent, authorization, or legal right\n\n   5. Representing that the use of Llama 3.3 or outputs are human-generated\n\n   6. Generating or facilitating false online engagement, including fake reviews and other means of fake online engagement\n\n4. Fail to appropriately disclose to end users any known dangers of your AI system\n5. Interact with third party tools, models, or software designed to generate unlawful content or engage in unlawful or harmful conduct and/or represent that the outputs of such tools, models, or software are associated with Meta or Llama 3.3\nWith respect to any multimodal models included in Llama 3.3, the rights granted under Section 1(a) of the Llama 3.3 Community License Agreement are not being granted to you if you are an individual domiciled in, or a company with a principal place of business in, the European Union. This restriction does not apply to end users of a product or service that incorporates any such multimodal models.\nPlease report any violation of this Policy, software “bug,” or other problems that could lead to a violation of this Policy through one of the following means:\n* Reporting issues with the model: [https://github.com/meta-llama/llama-models/issues](https://l.workplace.com/l.php?u=https%3A%2F%2Fgithub.com%2Fmeta-llama%2Fllama-models%2Fissues&h=AT0qV8W9BFT6NwihiOHRuKYQM_UnkzN_NmHMy91OT55gkLpgi4kQupHUl0ssR4dQsIQ8n3tfd0vtkobvsEvt1l4Ic6GXI2EeuHV8N08OG2WnbAmm0FL4ObkazC6G_256vN0lN9DsykCvCqGZ)   * Reporting risky content generated by the model: [developers.facebook.com/llama\\_output\\_feedback](http://developers.facebook.com/llama_output_feedback)   * Reporting bugs and security concerns: [facebook.com/whitehat/info](http://facebook.com/whitehat/info)   * Reporting violations of the Acceptable Use Policy or unlicensed uses of Llama 3.3: LlamaUseReport@meta.com  "
+extra_gated_fields:
+  First Name: text
+  Last Name: text
+  Date of birth: date_picker
+  Country: country
+  Affiliation: text
+  Job title:
+    type: select
+    options:
+    - Student
+    - Research Graduate
+    - AI researcher
+    - AI developer/engineer
+    - Reporter
+    - Other
+  geo: ip_location
+  By clicking Submit below I accept the terms of the license and acknowledge that the information I provide will be collected stored processed and shared in accordance with the Meta Privacy Policy: checkbox
+extra_gated_description: >-
+  The information you provide will be collected, stored, processed and shared in
+  accordance with the [Meta Privacy
+  Policy](https://www.facebook.com/privacy/policy/).
+extra_gated_button_content: Submit
+license: llama3.3
+---
+## Model Information
+
+The Meta Llama 3.3 multilingual large language model (LLM) is an instruction tuned generative model in 70B (text in/text out). The Llama 3.3 instruction tuned text only model is optimized for multilingual dialogue use cases and outperforms many of the available open source and closed chat models on common industry benchmarks.
+
+**Model developer**: Meta
+
+**Model Architecture:** Llama 3.3 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety. 
+
+|  | Training Data | Params | Input modalities | Output modalities | Context length | GQA | Token count | Knowledge cutoff |
+| :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- |
+| Llama 3.3 (text only)  | A new mix of publicly available online data. | 70B | Multilingual Text | Multilingual Text and code  | 128k | Yes | 15T+ | December 2023 |
+
+**Supported languages:** English, German, French, Italian, Portuguese, Hindi, Spanish, and Thai.
+
+**Llama 3.3 model**. Token counts refer to pretraining data only. All model versions use Grouped-Query Attention (GQA) for improved inference scalability.
+
+**Model Release Date:** 
+
+* **70B Instruct: December 6, 2024** 
+
+**Status:** This is a static model trained on an offline dataset. Future versions of the tuned models will be released as we improve model safety with community feedback.
+
+**License** A custom commercial license, the Llama 3.3 Community License Agreement, is available at: [https://github.com/meta-llama/llama-models/blob/main/models/llama3\_3/LICENSE](https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/LICENSE)
+
+Where to send questions or comments about the model Instructions on how to provide feedback or comments on the model can be found in the model [README](https://github.com/meta-llama/llama3). For more technical information about generation parameters and recipes for how to use Llama 3.3 in applications, please go [here](https://github.com/meta-llama/llama-recipes). 
+
+## Intended Use
+
+**Intended Use Cases** Llama 3.3 is intended for commercial and research use in multiple languages. Instruction tuned text only models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks. The Llama 3.3 model also supports the ability to leverage the outputs of its models to improve other models including synthetic data generation and distillation. The Llama 3.3 Community License allows for these use cases. 
+
+**Out-of-scope** Use in any manner that violates applicable laws or regulations (including trade compliance laws). Use in any other way that is prohibited by the Acceptable Use Policy and Llama 3.3 Community License. Use in languages beyond those explicitly referenced as supported in this model card\*\*.
+
+\*\*Note: Llama 3.3 has been trained on a broader collection of languages than the 8 supported languages. Developers may fine-tune Llama 3.3 models for languages beyond the 8 supported languages provided they comply with the Llama 3.3 Community License and the Acceptable Use Policy and in such cases are responsible for ensuring that any uses of Llama 3.3 in additional languages is done in a safe and responsible manner.
+
+## How to use
+
+This repository contains two versions of Llama-3.3-70B-Instruct, for use with transformers and with the original `llama` codebase.
+
+### Use with transformers
+
+Starting with `transformers >= 4.45.0` onward, you can run conversational inference using the Transformers `pipeline` abstraction or by leveraging the Auto classes with the `generate()` function.
+
+Make sure to update your transformers installation via `pip install --upgrade transformers`.
+
+See the snippet below for usage with Transformers:
+
+```python
+import transformers
+import torch
+
+model_id = "meta-llama/Llama-3.3-70B-Instruct"
+
+pipeline = transformers.pipeline(
+    "text-generation",
+    model=model_id,
+    model_kwargs={"torch_dtype": torch.bfloat16},
+    device_map="auto",
+)
+
+messages = [
+    {"role": "system", "content": "You are a pirate chatbot who always responds in pirate speak!"},
+    {"role": "user", "content": "Who are you?"},
+]
+
+outputs = pipeline(
+    messages,
+    max_new_tokens=256,
+)
+print(outputs[0]["generated_text"][-1])
+```
+
+### Tool use with transformers
+
+LLaMA-3.3 supports multiple tool use formats. You can see a full guide to prompt formatting [here](https://llama.meta.com/docs/model-cards-and-prompt-formats/llama3_1/).
+
+Tool use is also supported through [chat templates](https://huggingface.co/docs/transformers/main/chat_templating#advanced-tool-use--function-calling) in Transformers. 
+Here is a quick example showing a single simple tool:
+
+```python
+# First, define a tool
+def get_current_temperature(location: str) -> float:
+    """
+    Get the current temperature at a location.
+    
+    Args:
+        location: The location to get the temperature for, in the format "City, Country"
+    Returns:
+        The current temperature at the specified location in the specified units, as a float.
+    """
+    return 22.  # A real function should probably actually get the temperature!
+
+# Next, create a chat and apply the chat template
+messages = [
+  {"role": "system", "content": "You are a bot that responds to weather queries."},
+  {"role": "user", "content": "Hey, what's the temperature in Paris right now?"}
+]
+
+inputs = tokenizer.apply_chat_template(messages, tools=[get_current_temperature], add_generation_prompt=True)
+```
+
+You can then generate text from this input as normal. If the model generates a tool call, you should add it to the chat like so:
+
+```python
+tool_call = {"name": "get_current_temperature", "arguments": {"location": "Paris, France"}}
+messages.append({"role": "assistant", "tool_calls": [{"type": "function", "function": tool_call}]})
+```
+
+and then call the tool and append the result, with the `tool` role, like so:
+
+```python
+messages.append({"role": "tool", "name": "get_current_temperature", "content": "22.0"})
+```
+
+After that, you can `generate()` again to let the model use the tool result in the chat. Note that this was a very brief introduction to tool calling - for more information,
+see the [LLaMA prompt format docs](https://llama.meta.com/docs/model-cards-and-prompt-formats/llama3_1/) and the Transformers [tool use documentation](https://huggingface.co/docs/transformers/main/chat_templating#advanced-tool-use--function-calling).
+
+
+### Use with `bitsandbytes`
+
+The model checkpoints can be used in `8-bit` and `4-bit` for further memory optimisations using `bitsandbytes` and `transformers`
+
+See the snippet below for usage:
+
+```python
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model_id = "meta-llama/Llama-3.3-70B-Instruct"
+quantization_config = BitsAndBytesConfig(load_in_8bit=True)
+
+quantized_model = AutoModelForCausalLM.from_pretrained(
+	model_id, device_map="auto", torch_dtype=torch.bfloat16, quantization_config=quantization_config)
+
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+input_text = "What are we having for dinner?"
+input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")
+
+output = quantized_model.generate(**input_ids, max_new_tokens=10)
+
+print(tokenizer.decode(output[0], skip_special_tokens=True))
+```
+
+To load in 4-bit simply pass `load_in_4bit=True`
+
+### Use with `llama`
+
+Please, follow the instructions in the [repository](https://github.com/meta-llama/llama).
+
+To download Original checkpoints, see the example command below leveraging `huggingface-cli`:
+
+```
+huggingface-cli download meta-llama/Llama-3.3-70B-Instruct --include "original/*" --local-dir Llama-3.3-70B-Instruct
+```
+
+## Hardware and Software
+
+**Training Factors** We used custom training libraries, Meta's custom built GPU cluster, and production infrastructure for pretraining. Fine-tuning, annotation, and evaluation were also performed on production infrastructure.
+
+**Training Energy Use** Training utilized a cumulative of **39.3**M GPU hours of computation on H100-80GB (TDP of 700W) type hardware, per the table below. Training time is the total GPU time required for training each model and power consumption is the peak power capacity per GPU device used, adjusted for power usage efficiency. 
+
+## 
+
+## **Training Greenhouse Gas Emissions** Estimated total location-based greenhouse gas emissions were **11,390** tons CO2eq for training. Since 2020, Meta has maintained net zero greenhouse gas emissions in its global operations and matched 100% of its electricity use with renewable energy, therefore the total market-based greenhouse gas emissions for training were 0 tons CO2eq.
+
+|  | Training Time (GPU hours) | Training Power Consumption (W) | Training Location-Based Greenhouse Gas Emissions (tons CO2eq) | Training Market-Based Greenhouse Gas Emissions (tons CO2eq) |
+| :---- | :---: | :---: | :---: | :---: |
+| Llama 3.3 70B | 7.0M | 700 | 2,040 | 0 |
+
+## The methodology used to determine training energy use and greenhouse gas emissions can be found [here](https://arxiv.org/pdf/2204.05149).  Since Meta is openly releasing these models, the training energy use and greenhouse gas emissions  will not be incurred by others.
+
+## Training Data
+
+**Overview:** Llama 3.3 was pretrained on \~15 trillion tokens of data from publicly available sources. The fine-tuning data includes publicly available instruction datasets, as well as over 25M synthetically generated examples. 
+
+**Data Freshness:** The pretraining data has a cutoff of December 2023\.
+
+## Benchmarks \- English Text
+
+In this section, we report the results for Llama 3.3 relative to our previous models. 
+
+### Instruction tuned models
+
+## 
+
+| Category | Benchmark | \# Shots | Metric | Llama 3.1 8B Instruct | Llama 3.1 70B Instruct | Llama-3.3 70B Instruct | Llama 3.1 405B Instruct |
+| :---- | :---- | ----- | :---- | ----- | ----- | ----- | ----- |
+|  | MMLU (CoT) | 0 | macro\_avg/acc | 73.0 | 86.0 | 86.0 | 88.6 |
+|  | MMLU Pro (CoT) | 5 | macro\_avg/acc | 48.3 | 66.4 | 68.9 | 73.3 |
+| Steerability | IFEval |  |  | 80.4 | 87.5 | 92.1 | 88.6 |
+| Reasoning | GPQA Diamond (CoT) | 0 | acc | 31.8 | 48.0 | 50.5 | 49.0 |
+| Code | HumanEval | 0 | pass@1 | 72.6 | 80.5 | 88.4 | 89.0 |
+|  | MBPP EvalPlus (base) | 0 | pass@1 | 72.8 | 86.0 | 87.6 | 88.6 |
+| Math | MATH (CoT) | 0 | sympy\_intersection\_score | 51.9 | 68.0 | 77.0 | 73.8 |
+| Tool Use | BFCL v2 | 0 | overall\_ast\_summary/macro\_avg/valid | 65.4 | 77.5 | 77.3 | 81.1 |
+| Multilingual | MGSM | 0 | em | 68.9 | 86.9 | 91.1 | 91.6 |
+
+## 
+
+## Responsibility & Safety
+
+As part of our Responsible release approach, we followed a three-pronged strategy to managing trust & safety risks:
+
+* Enable developers to deploy helpful, safe and flexible experiences for their target audience and for the use cases supported by Llama.
+* Protect developers against adversarial users aiming to exploit Llama capabilities to potentially cause harm.
+* Provide protections for the community to help prevent the misuse of our models.
+
+### Responsible deployment
+
+Llama is a foundational technology designed to be used in a variety of use cases, examples on how Meta’s Llama models have been responsibly deployed can be found in our [Community Stories webpage](https://llama.meta.com/community-stories/). Our approach is to build the most helpful models enabling the world to benefit from the technology power, by aligning our model safety for the generic use cases addressing a standard set of harms. Developers are then in the driver seat to tailor safety for their use case, defining their own policy and deploying the models with the necessary safeguards in their Llama systems. Llama 3.3 was developed following the best practices outlined in our Responsible Use Guide, you can refer to the [Responsible Use Guide](https://llama.meta.com/responsible-use-guide/) to learn more.
+
+#### Llama 3.3 instruct
+
+Our main objectives for conducting safety fine-tuning are to provide the research community with a valuable resource for studying the robustness of safety fine-tuning, as well as to offer developers a readily available, safe, and powerful model for various applications to reduce the developer workload to deploy safe AI systems. For more details on the safety mitigations implemented please read the Llama 3 paper.
+
+**Fine-tuning data**
+We employ a multi-faceted approach to data collection, combining human-generated data from our vendors with synthetic data to mitigate potential safety risks. We’ve developed many large language model (LLM)-based classifiers that enable us to thoughtfully select high-quality prompts and responses, enhancing data quality control.
+
+**Refusals and Tone**
+Building on the work we started with Llama 3, we put a great emphasis on model refusals to benign prompts as well as refusal tone. We included both borderline and adversarial prompts in our safety data strategy, and modified our safety data responses to follow tone guidelines.
+
+#### Llama 3.3 systems
+
+**Large language models, including Llama 3.3, are not designed to be deployed in isolation but instead should be deployed as part of an overall AI system with additional safety guardrails as required.** Developers are expected to deploy system safeguards when building agentic systems. Safeguards are key to achieve the right helpfulness-safety alignment as well as mitigating safety and security risks inherent to the system and any integration of the model or system with external tools.
+As part of our responsible release approach, we provide the community with [safeguards](https://llama.meta.com/trust-and-safety/) that developers should deploy with Llama models or other LLMs, including Llama Guard 3, Prompt Guard and Code Shield. All our [reference implementations](https://github.com/meta-llama/llama-agentic-system) demos contain these safeguards by default so developers can benefit from system-level safety out-of-the-box.
+
+#### Capability specific considerations 
+
+**Tool-use**: Just like in standard software development, developers are responsible for the integration of the LLM with the tools and services of their choice. They should define a clear policy for their use case and assess the integrity of the third party services they use to be aware of the safety and security limitations when using this capability. Refer to the Responsible Use Guide for best practices on the safe deployment of the third party safeguards. 
+
+**Multilinguality**: Llama 3.3 supports 7 languages in addition to English: French, German, Hindi, Italian, Portuguese, Spanish, and Thai. Llama may be able to output text in other languages than those that meet performance thresholds for safety and helpfulness. We strongly discourage developers from using this model to converse in non-supported languages without implementing finetuning and system controls in alignment with their policies and the best practices shared in the Responsible Use Guide. 
+
+### Evaluations
+
+We evaluated Llama models for common use cases as well as specific capabilities. Common use cases evaluations measure safety risks of systems for most commonly built applications including chat bot, coding assistant, tool calls. We built dedicated, adversarial evaluation datasets and evaluated systems composed of Llama models and Llama Guard 3 to filter input prompt and output response. It is important to evaluate applications in context, and we recommend building dedicated evaluation dataset for your use case. Prompt Guard and Code Shield are also available if relevant to the application.   
+Capability evaluations measure vulnerabilities of Llama models inherent to specific capabilities, for which were crafted dedicated benchmarks including long context, multilingual, tools calls, coding or memorization.
+
+**Red teaming**   
+For both scenarios, we conducted recurring red teaming exercises with the goal of discovering risks via adversarial prompting and we used the learnings to improve our benchmarks and safety tuning datasets.   
+We partnered early with subject-matter experts in critical risk areas to understand the nature of these real-world harms and how such models may lead to unintended harm for society. Based on these conversations, we derived a set of adversarial goals for the red team to attempt to achieve, such as extracting harmful information or reprogramming the model to act in a potentially harmful capacity.  The red team consisted of experts in cybersecurity, adversarial machine learning, responsible AI, and integrity in addition to multilingual content specialists with background in integrity issues in specific geographic markets. . 
+
+### Critical and other risks 
+
+### We specifically focused our efforts on mitigating the following critical risk areas:
+
+**1- CBRNE (Chemical, Biological, Radiological, Nuclear, and Explosive materials) helpfulness**
+To assess risks related to proliferation of chemical and biological weapons of the Llama 3 family of models, we performed uplift testing designed to assess whether use of the Llama 3 models could meaningfully increase the capabilities of malicious actors to plan or carry out attacks using these types of weapons.
+
+### **2\. Child Safety**
+
+Child Safety risk assessments were conducted using a team of experts, to assess the model’s capability to produce outputs that could result in Child Safety risks and inform on any necessary and appropriate risk mitigations via fine tuning. We leveraged those expert red teaming sessions to expand the coverage of our evaluation benchmarks through Llama 3 model development.  For Llama 3, we conducted new in-depth sessions using objective based methodologies to assess the model risks along multiple attack vectors including the additional languages Llama 3 is trained on. We also partnered with content specialists to perform red teaming exercises assessing potentially violating content while taking account of market specific nuances or experiences. 
+
+**3\. Cyber attack enablement**
+Our cyber attack uplift study investigated whether the Llama 3 family of LLMs can enhance human capabilities in hacking tasks, both in terms of skill level and speed.
+Our attack automation study focused on evaluating the capabilities of LLMs when used as autonomous agents in cyber offensive operations, specifically in the context of ransomware attacks. This evaluation was distinct from previous studies that considered LLMs as interactive assistants. The primary objective was to assess whether these models could effectively function as independent agents in executing complex cyber-attacks without human intervention.
+
+### Community 
+
+Generative AI safety requires expertise and tooling, and we believe in the strength of the open community to accelerate its progress. We are active members of open consortiums, including the AI Alliance, Partnership on AI and MLCommons, actively contributing to safety standardization and transparency. We encourage the community to adopt taxonomies like the MLCommons Proof of Concept evaluation to facilitate collaboration and transparency on safety and content evaluations. Our Purple Llama tools are open sourced for the community to use and widely distributed across ecosystem partners including cloud service providers. We encourage community contributions to our [Github repository](https://github.com/meta-llama/PurpleLlama). 
+
+We also set up the [Llama Impact Grants](https://llama.meta.com/llama-impact-grants/) program to identify and support the most compelling applications of Meta’s Llama model for societal benefit across three categories: education, climate and open innovation. The 20 finalists from the hundreds of applications can be found [here](https://llama.meta.com/llama-impact-grants/#finalists). 
+
+Finally, we put in place a set of resources including an [output reporting mechanism](https://developers.facebook.com/llama_output_feedback) and [bug bounty program](https://www.facebook.com/whitehat) to continuously improve the Llama technology with the help of the community.
+
+## Ethical Considerations and Limitations
+
+The core values of Llama 3.3 are openness, inclusivity and helpfulness. It is meant to serve everyone, and to work for a wide range of use cases. It is thus designed to be accessible to people across many different backgrounds, experiences and perspectives. Llama 3.3 addresses users and their needs as they are, without insertion unnecessary judgment or normativity, while reflecting the understanding that even content that may appear problematic in some cases can serve valuable purposes in others. It respects the dignity and autonomy of all users, especially in terms of the values of free thought and expression that power innovation and progress. 
+
+But Llama 3.3 is a new technology, and like any new technology, there are risks associated with its use. Testing conducted to date has not covered, nor could it cover, all scenarios. For these reasons, as with all LLMs, Llama 3.3’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Llama 3.3 model, developers should perform safety testing and tuning tailored to their specific applications of the model. Please refer to available resources including our [Responsible Use Guide](https://llama.meta.com/responsible-use-guide), [Trust and Safety](https://llama.meta.com/trust-and-safety/) solutions, and other [resources](https://llama.meta.com/docs/get-started/) to learn more about responsible development.

USE_POLICY.md

@@ -0,0 +1,73 @@
+**Llama 3.3** **Acceptable Use Policy**
+
+Meta is committed to promoting safe and fair use of its tools and features, including Llama 3.3. If you access or use Llama 3.3, you agree to this Acceptable Use Policy (“**Policy**”). The most recent copy of this policy can be found at [https://www.llama.com/llama3\_3/use-policy](https://www.llama.com/llama3_3/use-policy).
+
+**Prohibited Uses**
+
+We want everyone to use Llama 3.3 safely and responsibly. You agree you will not use, or allow others to use, Llama 3.3 to:
+
+1. Violate the law or others’ rights, including to:
+
+   1. Engage in, promote, generate, contribute to, encourage, plan, incite, or further illegal or unlawful activity or content, such as:  
+      1. Violence or terrorism  
+      2. Exploitation or harm to children, including the solicitation, creation, acquisition, or dissemination of child exploitative content or failure to report Child Sexual Abuse Material  
+      3. Human trafficking, exploitation, and sexual violence  
+      4. The illegal distribution of information or materials to minors, including obscene materials, or failure to employ legally required age-gating in connection with such information or materials.  
+      5. Sexual solicitation  
+      6. Any other criminal activity
+
+   2. Engage in, promote, incite, or facilitate the harassment, abuse, threatening, or bullying of individuals or groups of individuals
+
+   3. Engage in, promote, incite, or facilitate discrimination or other unlawful or harmful conduct in the provision of employment, employment benefits, credit, housing, other economic benefits, or other essential goods and services
+
+   4. Engage in the unauthorized or unlicensed practice of any profession including, but not limited to, financial, legal, medical/health, or related professional practices
+
+   5. Collect, process, disclose, generate, or infer private or sensitive information about individuals, including information about individuals’ identity, health, or demographic information, unless you have obtained the right to do so in accordance with applicable law
+
+   6. Engage in or facilitate any action or generate any content that infringes, misappropriates, or otherwise violates any third-party rights, including the outputs or results of any products or services using the Llama Materials
+
+   7. Create, generate, or facilitate the creation of malicious code, malware, computer viruses or do anything else that could disable, overburden, interfere with or impair the proper working, integrity, operation or appearance of a website or computer system
+
+   8. Engage in any action, or facilitate any action, to intentionally circumvent or remove usage restrictions or other safety measures, or to enable functionality disabled by Meta 
+
+2. Engage in, promote, incite, facilitate, or assist in the planning or development of activities that present a risk of death or bodily harm to individuals, including use of Llama 3.3 related to the following:
+
+   1. Military, warfare, nuclear industries or applications, espionage, use for materials or activities that are subject to the International Traffic Arms Regulations (ITAR) maintained by the United States Department of State or to the U.S. Biological Weapons Anti-Terrorism Act of 1989 or the Chemical Weapons Convention Implementation Act of 1997
+
+   2. Guns and illegal weapons (including weapon development)
+
+   3. Illegal drugs and regulated/controlled substances
+
+   4. Operation of critical infrastructure, transportation technologies, or heavy machinery
+
+   5. Self-harm or harm to others, including suicide, cutting, and eating disorders
+
+   6. Any content intended to incite or promote violence, abuse, or any infliction of bodily harm to an individual
+
+3. Intentionally deceive or mislead others, including use of Llama 3.3 related to the following:
+
+   1. Generating, promoting, or furthering fraud or the creation or promotion of disinformation
+
+   2. Generating, promoting, or furthering defamatory content, including the creation of defamatory statements, images, or other content
+
+   3. Generating, promoting, or further distributing spam
+
+   4. Impersonating another individual without consent, authorization, or legal right
+
+   5. Representing that the use of Llama 3.3 or outputs are human-generated
+
+   6. Generating or facilitating false online engagement, including fake reviews and other means of fake online engagement 
+
+4. Fail to appropriately disclose to end users any known dangers of your AI system
+
+5. Interact with third party tools, models, or software designed to generate unlawful content or engage in unlawful or harmful conduct and/or represent that the outputs of such tools, models, or software are associated with Meta or Llama 3.3
+
+With respect to any multimodal models included in Llama 3.3, the rights granted under Section 1(a) of the Llama 3.3 Community License Agreement are not being granted to you if you are an individual domiciled in, or a company with a principal place of business in, the European Union. This restriction does not apply to end users of a product or service that incorporates any such multimodal models.
+
+Please report any violation of this Policy, software “bug,” or other problems that could lead to a violation of this Policy through one of the following means:
+
+* Reporting issues with the model: [https://github.com/meta-llama/llama-models/issues](https://l.workplace.com/l.php?u=https%3A%2F%2Fgithub.com%2Fmeta-llama%2Fllama-models%2Fissues&h=AT0qV8W9BFT6NwihiOHRuKYQM_UnkzN_NmHMy91OT55gkLpgi4kQupHUl0ssR4dQsIQ8n3tfd0vtkobvsEvt1l4Ic6GXI2EeuHV8N08OG2WnbAmm0FL4ObkazC6G_256vN0lN9DsykCvCqGZ)  
+* Reporting risky content generated by the model: [developers.facebook.com/llama\_output\_feedback](http://developers.facebook.com/llama_output_feedback)  
+* Reporting bugs and security concerns: [facebook.com/whitehat/info](http://facebook.com/whitehat/info)  
+* Reporting violations of the Acceptable Use Policy or unlicensed uses of Llama 3.3: LlamaUseReport@meta.com
+

config.json

@@ -0,0 +1,39 @@
+{
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 8192,
+  "initializer_range": 0.02,
+  "intermediate_size": 28672,
+  "max_position_embeddings": 131072,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 64,
+  "num_hidden_layers": 80,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "factor": 8.0,
+    "high_freq_factor": 4.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "llama3"
+  },
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.47.0.dev0",
+  "use_cache": true,
+  "vocab_size": 128256
+}

generation_config.json

@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 128000,
+  "do_sample": true,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.47.0.dev0"
+}

model-00002-of-00030.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fcd250bf27cc5c51c2661eeb98f4d2de5525c6d466a0ee690bc80e29a917bcd4
+size 4664167376

model-00005-of-00030.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56e1ceef3622bef0a91b554a8dba1074b93f4f0eb1d28c38df112672fd31cb3f
+size 4664134408

model-00006-of-00030.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7662172fb0dd22f764869236f0a42c6f79fb385e5bced923936292cb8d94bb56
+size 4664167408

model-00007-of-00030.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0c90f7ff73f765286f51799f512a8c1cfeb4f0a46c4f2831fc855dd7ee0a8b1
+size 4664167408

model-00011-of-00030.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:458aeafc39b1c137d7d11df3de45f2d9db3175363230e9290ec08d6531fcdd26
+size 4664167408

model-00012-of-00030.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:962683d58eb1aa7b7efa35451fbf21d605ab8eb718d854766db0cfb6bfca1f5a
+size 4664167408

model-00015-of-00030.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6622c84f6c07c1ca73dabbbc63c7e7842d98ba1a92bbe18d9f485718b12bf757
+size 4664134408

model-00016-of-00030.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87dd8ffa1c6fbefefac8bfd02e7c3001f1c6b31e2afe9859038e912b6712f884
+size 4664167408

model-00017-of-00030.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:302f92d01ba80999d73b93f4e4d90911b260e7e78c8e61605f43a775a77cadf8
+size 4664167408

model-00021-of-00030.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21097441bb201b3094daf3601a6fc9d98e3b62b9cd76182ff00d39de5c16bf93
+size 4664167408

model-00022-of-00030.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9f2e5a40d0f563164874ceacb7f785481a908fc8fb46321217f39a6b9d2bad3f
+size 4664167408

model-00027-of-00030.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddff9ad869a4dd4df1d361cdd0f382937877908ded87da41b8bdf89f17ce64f3
+size 4664167408

model-00030-of-00030.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0436c98c9b4f78e61f6769e588e0853179b970e2a084ae2886e2daebf1ce13d
+size 2101346432

model.safetensors.index.json

@@ -0,0 +1,730 @@
+{
+  "metadata": {
+    "total_size": 141107412992
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00030-of-00030.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00030.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00030.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00030.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00030.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00030.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00030.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00030.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00030.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00030.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00030.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00002-of-00030.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00030.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00002-of-00030.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00030.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00030.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00030.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00030.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00005-of-00030.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00005-of-00030.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00005-of-00030.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00005-of-00030.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00006-of-00030.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00006-of-00030.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00005-of-00030.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00006-of-00030.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00006-of-00030.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00006-of-00030.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00006-of-00030.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00007-of-00030.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00007-of-00030.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00006-of-00030.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00007-of-00030.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00007-of-00030.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00007-of-00030.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00007-of-00030.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00008-of-00030.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00008-of-00030.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00007-of-00030.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00008-of-00030.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00008-of-00030.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00002-of-00030.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00030.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00008-of-00030.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00008-of-00030.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00009-of-00030.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00009-of-00030.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00008-of-00030.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00009-of-00030.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00009-of-00030.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00009-of-00030.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00009-of-00030.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00009-of-00030.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00010-of-00030.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00010-of-00030.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00010-of-00030.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00010-of-00030.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00011-of-00030.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00011-of-00030.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00010-of-00030.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00011-of-00030.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00011-of-00030.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00011-of-00030.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00011-of-00030.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00012-of-00030.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00012-of-00030.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00011-of-00030.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00002-of-00030.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00030.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00012-of-00030.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00012-of-00030.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00012-of-00030.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00012-of-00030.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00013-of-00030.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00013-of-00030.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00012-of-00030.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00013-of-00030.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00013-of-00030.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00013-of-00030.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00013-of-00030.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00014-of-00030.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00014-of-00030.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00013-of-00030.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00014-of-00030.safetensors",
+    "model.layers.36.mlp.down_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.36.mlp.gate_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.36.mlp.up_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00014-of-00030.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00014-of-00030.safetensors",
+    "model.layers.37.mlp.down_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.37.mlp.gate_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.37.mlp.up_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00014-of-00030.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00014-of-00030.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00015-of-00030.safetensors",
+    "model.layers.38.mlp.down_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.38.mlp.gate_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.38.mlp.up_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00015-of-00030.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00015-of-00030.safetensors",
+    "model.layers.39.mlp.down_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.39.mlp.gate_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.39.mlp.up_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00015-of-00030.safetensors",
+    "model.layers.39.self_attn.k_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.39.self_attn.q_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.39.self_attn.v_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00003-of-00030.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00003-of-00030.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00030.safetensors",
+    "model.layers.40.input_layernorm.weight": "model-00016-of-00030.safetensors",
+    "model.layers.40.mlp.down_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.40.mlp.gate_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.40.mlp.up_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.40.post_attention_layernorm.weight": "model-00016-of-00030.safetensors",
+    "model.layers.40.self_attn.k_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.40.self_attn.o_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.40.self_attn.q_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.40.self_attn.v_proj.weight": "model-00015-of-00030.safetensors",
+    "model.layers.41.input_layernorm.weight": "model-00016-of-00030.safetensors",
+    "model.layers.41.mlp.down_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.41.mlp.gate_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.41.mlp.up_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.41.post_attention_layernorm.weight": "model-00016-of-00030.safetensors",
+    "model.layers.41.self_attn.k_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.41.self_attn.o_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.41.self_attn.q_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.41.self_attn.v_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.42.input_layernorm.weight": "model-00016-of-00030.safetensors",
+    "model.layers.42.mlp.down_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.42.mlp.gate_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.42.mlp.up_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.42.post_attention_layernorm.weight": "model-00016-of-00030.safetensors",
+    "model.layers.42.self_attn.k_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.42.self_attn.o_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.42.self_attn.q_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.42.self_attn.v_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.43.input_layernorm.weight": "model-00017-of-00030.safetensors",
+    "model.layers.43.mlp.down_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.43.mlp.gate_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.43.mlp.up_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.43.post_attention_layernorm.weight": "model-00017-of-00030.safetensors",
+    "model.layers.43.self_attn.k_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.43.self_attn.o_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.43.self_attn.q_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.43.self_attn.v_proj.weight": "model-00016-of-00030.safetensors",
+    "model.layers.44.input_layernorm.weight": "model-00017-of-00030.safetensors",
+    "model.layers.44.mlp.down_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.44.mlp.gate_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.44.mlp.up_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.44.post_attention_layernorm.weight": "model-00017-of-00030.safetensors",
+    "model.layers.44.self_attn.k_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.44.self_attn.o_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.44.self_attn.q_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.44.self_attn.v_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.45.input_layernorm.weight": "model-00017-of-00030.safetensors",
+    "model.layers.45.mlp.down_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.45.mlp.gate_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.45.mlp.up_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.45.post_attention_layernorm.weight": "model-00017-of-00030.safetensors",
+    "model.layers.45.self_attn.k_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.45.self_attn.o_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.45.self_attn.q_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.45.self_attn.v_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.46.input_layernorm.weight": "model-00018-of-00030.safetensors",
+    "model.layers.46.mlp.down_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.46.mlp.gate_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.46.mlp.up_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.46.post_attention_layernorm.weight": "model-00018-of-00030.safetensors",
+    "model.layers.46.self_attn.k_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.46.self_attn.o_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.46.self_attn.q_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.46.self_attn.v_proj.weight": "model-00017-of-00030.safetensors",
+    "model.layers.47.input_layernorm.weight": "model-00018-of-00030.safetensors",
+    "model.layers.47.mlp.down_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.47.mlp.gate_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.47.mlp.up_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.47.post_attention_layernorm.weight": "model-00018-of-00030.safetensors",
+    "model.layers.47.self_attn.k_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.47.self_attn.o_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.47.self_attn.q_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.47.self_attn.v_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.48.input_layernorm.weight": "model-00018-of-00030.safetensors",
+    "model.layers.48.mlp.down_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.48.mlp.gate_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.48.mlp.up_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.48.post_attention_layernorm.weight": "model-00018-of-00030.safetensors",
+    "model.layers.48.self_attn.k_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.48.self_attn.o_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.48.self_attn.q_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.48.self_attn.v_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.49.input_layernorm.weight": "model-00019-of-00030.safetensors",
+    "model.layers.49.mlp.down_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.49.mlp.gate_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.49.mlp.up_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.49.post_attention_layernorm.weight": "model-00019-of-00030.safetensors",
+    "model.layers.49.self_attn.k_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.49.self_attn.o_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.49.self_attn.q_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.49.self_attn.v_proj.weight": "model-00018-of-00030.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00003-of-00030.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00003-of-00030.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.50.input_layernorm.weight": "model-00019-of-00030.safetensors",
+    "model.layers.50.mlp.down_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.50.mlp.gate_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.50.mlp.up_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.50.post_attention_layernorm.weight": "model-00019-of-00030.safetensors",
+    "model.layers.50.self_attn.k_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.50.self_attn.o_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.50.self_attn.q_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.50.self_attn.v_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.51.input_layernorm.weight": "model-00019-of-00030.safetensors",
+    "model.layers.51.mlp.down_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.51.mlp.gate_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.51.mlp.up_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.51.post_attention_layernorm.weight": "model-00019-of-00030.safetensors",
+    "model.layers.51.self_attn.k_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.51.self_attn.o_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.51.self_attn.q_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.51.self_attn.v_proj.weight": "model-00019-of-00030.safetensors",
+    "model.layers.52.input_layernorm.weight": "model-00020-of-00030.safetensors",
+    "model.layers.52.mlp.down_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.52.mlp.gate_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.52.mlp.up_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.52.post_attention_layernorm.weight": "model-00020-of-00030.safetensors",
+    "model.layers.52.self_attn.k_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.52.self_attn.o_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.52.self_attn.q_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.52.self_attn.v_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.53.input_layernorm.weight": "model-00020-of-00030.safetensors",
+    "model.layers.53.mlp.down_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.53.mlp.gate_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.53.mlp.up_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.53.post_attention_layernorm.weight": "model-00020-of-00030.safetensors",
+    "model.layers.53.self_attn.k_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.53.self_attn.o_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.53.self_attn.q_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.53.self_attn.v_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.54.input_layernorm.weight": "model-00021-of-00030.safetensors",
+    "model.layers.54.mlp.down_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.54.mlp.gate_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.54.mlp.up_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.54.post_attention_layernorm.weight": "model-00021-of-00030.safetensors",
+    "model.layers.54.self_attn.k_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.54.self_attn.o_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.54.self_attn.q_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.54.self_attn.v_proj.weight": "model-00020-of-00030.safetensors",
+    "model.layers.55.input_layernorm.weight": "model-00021-of-00030.safetensors",
+    "model.layers.55.mlp.down_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.55.mlp.gate_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.55.mlp.up_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.55.post_attention_layernorm.weight": "model-00021-of-00030.safetensors",
+    "model.layers.55.self_attn.k_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.55.self_attn.o_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.55.self_attn.q_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.55.self_attn.v_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.56.input_layernorm.weight": "model-00021-of-00030.safetensors",
+    "model.layers.56.mlp.down_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.56.mlp.gate_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.56.mlp.up_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.56.post_attention_layernorm.weight": "model-00021-of-00030.safetensors",
+    "model.layers.56.self_attn.k_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.56.self_attn.o_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.56.self_attn.q_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.56.self_attn.v_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.57.input_layernorm.weight": "model-00022-of-00030.safetensors",
+    "model.layers.57.mlp.down_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.57.mlp.gate_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.57.mlp.up_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.57.post_attention_layernorm.weight": "model-00022-of-00030.safetensors",
+    "model.layers.57.self_attn.k_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.57.self_attn.o_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.57.self_attn.q_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.57.self_attn.v_proj.weight": "model-00021-of-00030.safetensors",
+    "model.layers.58.input_layernorm.weight": "model-00022-of-00030.safetensors",
+    "model.layers.58.mlp.down_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.58.mlp.gate_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.58.mlp.up_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.58.post_attention_layernorm.weight": "model-00022-of-00030.safetensors",
+    "model.layers.58.self_attn.k_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.58.self_attn.o_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.58.self_attn.q_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.58.self_attn.v_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.59.input_layernorm.weight": "model-00022-of-00030.safetensors",
+    "model.layers.59.mlp.down_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.59.mlp.gate_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.59.mlp.up_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.59.post_attention_layernorm.weight": "model-00022-of-00030.safetensors",
+    "model.layers.59.self_attn.k_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.59.self_attn.o_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.59.self_attn.q_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.59.self_attn.v_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00003-of-00030.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00003-of-00030.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.60.input_layernorm.weight": "model-00023-of-00030.safetensors",
+    "model.layers.60.mlp.down_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.60.mlp.gate_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.60.mlp.up_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.60.post_attention_layernorm.weight": "model-00023-of-00030.safetensors",
+    "model.layers.60.self_attn.k_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.60.self_attn.o_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.60.self_attn.q_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.60.self_attn.v_proj.weight": "model-00022-of-00030.safetensors",
+    "model.layers.61.input_layernorm.weight": "model-00023-of-00030.safetensors",
+    "model.layers.61.mlp.down_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.61.mlp.gate_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.61.mlp.up_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.61.post_attention_layernorm.weight": "model-00023-of-00030.safetensors",
+    "model.layers.61.self_attn.k_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.61.self_attn.o_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.61.self_attn.q_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.61.self_attn.v_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.62.input_layernorm.weight": "model-00023-of-00030.safetensors",
+    "model.layers.62.mlp.down_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.62.mlp.gate_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.62.mlp.up_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.62.post_attention_layernorm.weight": "model-00023-of-00030.safetensors",
+    "model.layers.62.self_attn.k_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.62.self_attn.o_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.62.self_attn.q_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.62.self_attn.v_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.63.input_layernorm.weight": "model-00024-of-00030.safetensors",
+    "model.layers.63.mlp.down_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.63.mlp.gate_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.63.mlp.up_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.63.post_attention_layernorm.weight": "model-00024-of-00030.safetensors",
+    "model.layers.63.self_attn.k_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.63.self_attn.o_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.63.self_attn.q_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.63.self_attn.v_proj.weight": "model-00023-of-00030.safetensors",
+    "model.layers.64.input_layernorm.weight": "model-00024-of-00030.safetensors",
+    "model.layers.64.mlp.down_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.64.mlp.gate_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.64.mlp.up_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.64.post_attention_layernorm.weight": "model-00024-of-00030.safetensors",
+    "model.layers.64.self_attn.k_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.64.self_attn.o_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.64.self_attn.q_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.64.self_attn.v_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.65.input_layernorm.weight": "model-00024-of-00030.safetensors",
+    "model.layers.65.mlp.down_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.65.mlp.gate_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.65.mlp.up_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.65.post_attention_layernorm.weight": "model-00024-of-00030.safetensors",
+    "model.layers.65.self_attn.k_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.65.self_attn.o_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.65.self_attn.q_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.65.self_attn.v_proj.weight": "model-00024-of-00030.safetensors",
+    "model.layers.66.input_layernorm.weight": "model-00025-of-00030.safetensors",
+    "model.layers.66.mlp.down_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.66.mlp.gate_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.66.mlp.up_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.66.post_attention_layernorm.weight": "model-00025-of-00030.safetensors",
+    "model.layers.66.self_attn.k_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.66.self_attn.o_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.66.self_attn.q_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.66.self_attn.v_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.67.input_layernorm.weight": "model-00025-of-00030.safetensors",
+    "model.layers.67.mlp.down_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.67.mlp.gate_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.67.mlp.up_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.67.post_attention_layernorm.weight": "model-00025-of-00030.safetensors",
+    "model.layers.67.self_attn.k_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.67.self_attn.o_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.67.self_attn.q_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.67.self_attn.v_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.68.input_layernorm.weight": "model-00026-of-00030.safetensors",
+    "model.layers.68.mlp.down_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.68.mlp.gate_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.68.mlp.up_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.68.post_attention_layernorm.weight": "model-00026-of-00030.safetensors",
+    "model.layers.68.self_attn.k_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.68.self_attn.o_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.68.self_attn.q_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.68.self_attn.v_proj.weight": "model-00025-of-00030.safetensors",
+    "model.layers.69.input_layernorm.weight": "model-00026-of-00030.safetensors",
+    "model.layers.69.mlp.down_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.69.mlp.gate_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.69.mlp.up_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.69.post_attention_layernorm.weight": "model-00026-of-00030.safetensors",
+    "model.layers.69.self_attn.k_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.69.self_attn.o_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.69.self_attn.q_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.69.self_attn.v_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00004-of-00030.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00004-of-00030.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00003-of-00030.safetensors",
+    "model.layers.70.input_layernorm.weight": "model-00026-of-00030.safetensors",
+    "model.layers.70.mlp.down_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.70.mlp.gate_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.70.mlp.up_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.70.post_attention_layernorm.weight": "model-00026-of-00030.safetensors",
+    "model.layers.70.self_attn.k_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.70.self_attn.o_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.70.self_attn.q_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.70.self_attn.v_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.71.input_layernorm.weight": "model-00027-of-00030.safetensors",
+    "model.layers.71.mlp.down_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.71.mlp.gate_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.71.mlp.up_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.71.post_attention_layernorm.weight": "model-00027-of-00030.safetensors",
+    "model.layers.71.self_attn.k_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.71.self_attn.o_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.71.self_attn.q_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.71.self_attn.v_proj.weight": "model-00026-of-00030.safetensors",
+    "model.layers.72.input_layernorm.weight": "model-00027-of-00030.safetensors",
+    "model.layers.72.mlp.down_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.72.mlp.gate_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.72.mlp.up_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.72.post_attention_layernorm.weight": "model-00027-of-00030.safetensors",
+    "model.layers.72.self_attn.k_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.72.self_attn.o_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.72.self_attn.q_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.72.self_attn.v_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.73.input_layernorm.weight": "model-00027-of-00030.safetensors",
+    "model.layers.73.mlp.down_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.73.mlp.gate_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.73.mlp.up_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.73.post_attention_layernorm.weight": "model-00027-of-00030.safetensors",
+    "model.layers.73.self_attn.k_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.73.self_attn.o_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.73.self_attn.q_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.73.self_attn.v_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.74.input_layernorm.weight": "model-00028-of-00030.safetensors",
+    "model.layers.74.mlp.down_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.74.mlp.gate_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.74.mlp.up_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.74.post_attention_layernorm.weight": "model-00028-of-00030.safetensors",
+    "model.layers.74.self_attn.k_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.74.self_attn.o_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.74.self_attn.q_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.74.self_attn.v_proj.weight": "model-00027-of-00030.safetensors",
+    "model.layers.75.input_layernorm.weight": "model-00028-of-00030.safetensors",
+    "model.layers.75.mlp.down_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.75.mlp.gate_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.75.mlp.up_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.75.post_attention_layernorm.weight": "model-00028-of-00030.safetensors",
+    "model.layers.75.self_attn.k_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.75.self_attn.o_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.75.self_attn.q_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.75.self_attn.v_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.76.input_layernorm.weight": "model-00028-of-00030.safetensors",
+    "model.layers.76.mlp.down_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.76.mlp.gate_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.76.mlp.up_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.76.post_attention_layernorm.weight": "model-00028-of-00030.safetensors",
+    "model.layers.76.self_attn.k_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.76.self_attn.o_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.76.self_attn.q_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.76.self_attn.v_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.77.input_layernorm.weight": "model-00029-of-00030.safetensors",
+    "model.layers.77.mlp.down_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.77.mlp.gate_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.77.mlp.up_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.77.post_attention_layernorm.weight": "model-00029-of-00030.safetensors",
+    "model.layers.77.self_attn.k_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.77.self_attn.o_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.77.self_attn.q_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.77.self_attn.v_proj.weight": "model-00028-of-00030.safetensors",
+    "model.layers.78.input_layernorm.weight": "model-00029-of-00030.safetensors",
+    "model.layers.78.mlp.down_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.78.mlp.gate_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.78.mlp.up_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.78.post_attention_layernorm.weight": "model-00029-of-00030.safetensors",
+    "model.layers.78.self_attn.k_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.78.self_attn.o_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.78.self_attn.q_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.78.self_attn.v_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.79.input_layernorm.weight": "model-00029-of-00030.safetensors",
+    "model.layers.79.mlp.down_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.79.mlp.gate_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.79.mlp.up_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.79.post_attention_layernorm.weight": "model-00029-of-00030.safetensors",
+    "model.layers.79.self_attn.k_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.79.self_attn.o_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.79.self_attn.q_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.79.self_attn.v_proj.weight": "model-00029-of-00030.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00004-of-00030.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00004-of-00030.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00004-of-00030.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00004-of-00030.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00004-of-00030.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00004-of-00030.safetensors",
+    "model.norm.weight": "model-00029-of-00030.safetensors"
+  }
+}

original/.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

original/README.md

@@ -0,0 +1,11 @@
+
+> [!IMPORTANT]  
+> This repository is a early-access checkpoint for Llama 3.1 70B.
+> This repo contains only Meta provided original checkpoints. Hugging Face checkpoints are available here.
+
+```bash
+You can invoke them via torchrun by doing the following:
+CHECKPOINT_DIR=~/.llama/checkpoints/Llama3.1-70B-Instruct-2014-12/
+pip install torch fairscale
+torchrun --nproc_per_node 8 `which example_chat_completion` "$CHECKPOINT_DIR"
+```

original/checklist.chk

@@ -0,0 +1,10 @@
+5ca57571675ec096c1918c76326e31ab  consolidated.00.pth
+2cb56d8cbe9aaf26552061e1602d6dc3  consolidated.01.pth
+90158e90c25112dfe037b94ad4779a8e  consolidated.02.pth
+201b59b7fe63f0d7697fdab3de8b5ab3  consolidated.03.pth
+716740eb1c34e0a221192c871529c4f6  consolidated.04.pth
+fa0787c90d69670a8d72084f37c264d0  consolidated.05.pth
+04d0acf24058c22ea13279276fba3a23  consolidated.06.pth
+1c0326797ca6a3c5d91c135e37257cec  consolidated.07.pth
+0956af3a2c289b9d3005f4bc559d6bb8  params.json
+08292403f8b173e7524d7fba7bbbd2d3  tokenizer.model

original/params.json

@@ -0,0 +1,12 @@
+{
+  "dim": 8192,
+  "n_layers": 80,
+  "n_heads": 64,
+  "n_kv_heads": 8,
+  "vocab_size": 128256,
+  "ffn_dim_multiplier": 1.3,
+  "multiple_of": 4096,
+  "norm_eps": 1e-05,
+  "rope_theta": 500000.0,
+  "use_scaled_rope": true
+}

original/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82e9d31979e92ab929cd544440f129d9ecd797b69e327f80f17e1c50d5551b55
+size 2183982

quant/g3/inc_output_hooks_maxabs_0_4.json

@@ -0,0 +1 @@
+{"GlobalRank": null, "LocalRank": 0, "Mode": "DynamicRange", "Nodes": {"model.layers.0.self_attn.qkv_proj": {"inputs": [[[4.09375]]], "params": {"weight": [[73.5]]}}, "model.layers.0.self_attn.o_proj": {"inputs": [[[0.1201171875]]], "outputs": [[[0.140625]], [[0.248046875]]], "params": {"weight": [[0.443359375]]}}, "model.layers.0.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.251953125]], [[5.53125]]]}, "model.layers.0.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.1513671875]]]}, "model.layers.0.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[576.0]]]}, "model.layers.0.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[124.0]]]}, "model.layers.0.self_attn.attn.impl.k_cache": {"inputs": [[[5.53125]]]}, "model.layers.0.self_attn.attn.impl.v_cache": {"inputs": [[[0.1513671875]]]}, "model.layers.0.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[3.015625]], [[5.53125]], [[0.1513671875]]], "outputs": [[[0.1123046875]], [[1.0]]]}, "model.layers.0.mlp.gate_up_proj": {"inputs": [[[1.453125]]], "params": {"weight": [[92.0]]}}, "model.layers.0.mlp.down_proj": {"inputs": [[[0.3203125]]], "outputs": [[[8.625]], [[52.5]]], "params": {"weight": [[0.83203125]]}}, "model.layers.1.self_attn.qkv_proj": {"inputs": [[[7.40625]]], "params": {"weight": [[10.625]]}}, "model.layers.1.self_attn.o_proj": {"inputs": [[[0.376953125]]], "outputs": [[[0.65625]], [[0.765625]]], "params": {"weight": [[0.52734375]]}}, "model.layers.1.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.74609375]], [[12.875]]]}, "model.layers.1.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.431640625]]]}, "model.layers.1.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[3.15625]]]}, "model.layers.1.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.0625]]]}, "model.layers.1.self_attn.attn.impl.k_cache": {"inputs": [[[12.875]]]}, "model.layers.1.self_attn.attn.impl.v_cache": {"inputs": [[[0.431640625]]]}, "model.layers.1.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.5625]], [[12.875]], [[0.421875]]], "outputs": [[[0.376953125]], [[1.0]]]}, "model.layers.1.mlp.gate_up_proj": {"inputs": [[[3.09375]]], "params": {"weight": [[4.46875]]}}, "model.layers.1.mlp.down_proj": {"inputs": [[[3.28125]]], "outputs": [[[2.171875]], [[5.0625]]], "params": {"weight": [[0.53125]]}}, "model.layers.2.self_attn.qkv_proj": {"inputs": [[[6.625]]], "params": {"weight": [[1.3828125]]}}, "model.layers.2.self_attn.o_proj": {"inputs": [[[0.07470703125]]], "outputs": [[[0.044189453125]], [[0.099609375]]], "params": {"weight": [[0.189453125]]}}, "model.layers.2.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.7578125]], [[11.4375]]]}, "model.layers.2.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.224609375]]]}, "model.layers.2.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.15625]]]}, "model.layers.2.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.328125]]]}, "model.layers.2.self_attn.attn.impl.k_cache": {"inputs": [[[11.4375]]]}, "model.layers.2.self_attn.attn.impl.v_cache": {"inputs": [[[0.224609375]]]}, "model.layers.2.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.3125]], [[11.0625]], [[0.2236328125]]], "outputs": [[[0.059814453125]], [[1.0]]]}, "model.layers.2.mlp.gate_up_proj": {"inputs": [[[2.578125]]], "params": {"weight": [[1.21875]]}}, "model.layers.2.mlp.down_proj": {"inputs": [[[0.58984375]]], "outputs": [[[0.345703125]], [[0.72265625]]], "params": {"weight": [[0.58984375]]}}, "model.layers.3.self_attn.qkv_proj": {"inputs": [[[3.796875]]], "params": {"weight": [[11.125]]}}, "model.layers.3.self_attn.o_proj": {"inputs": [[[0.1015625]]], "outputs": [[[0.64453125]], [[0.9453125]]], "params": {"weight": [[0.24609375]]}}, "model.layers.3.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5234375]], [[10.875]]]}, "model.layers.3.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.49609375]]]}, "model.layers.3.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.28125]]]}, "model.layers.3.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.515625]]]}, "model.layers.3.self_attn.attn.impl.k_cache": {"inputs": [[[10.875]]]}, "model.layers.3.self_attn.attn.impl.v_cache": {"inputs": [[[0.49609375]]]}, "model.layers.3.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.875]], [[10.6875]], [[0.49609375]]], "outputs": [[[0.1015625]], [[1.0]]]}, "model.layers.3.mlp.gate_up_proj": {"inputs": [[[22.625]]], "params": {"weight": [[14.4375]]}}, "model.layers.3.mlp.down_proj": {"inputs": [[[50.75]]], "outputs": [[[4.90625]], [[452.0]]], "params": {"weight": [[1.96875]]}}, "model.layers.4.self_attn.qkv_proj": {"inputs": [[[4.875]]], "params": {"weight": [[0.58984375]]}}, "model.layers.4.self_attn.o_proj": {"inputs": [[[0.1318359375]]], "outputs": [[[0.1015625]], [[0.154296875]]], "params": {"weight": [[0.39453125]]}}, "model.layers.4.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5546875]], [[16.0]]]}, "model.layers.4.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.82421875]]]}, "model.layers.4.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[3.4375]]]}, "model.layers.4.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.46875]]]}, "model.layers.4.self_attn.attn.impl.k_cache": {"inputs": [[[16.0]]]}, "model.layers.4.self_attn.attn.impl.v_cache": {"inputs": [[[0.82421875]]]}, "model.layers.4.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.0625]], [[15.5]], [[0.82421875]]], "outputs": [[[0.08056640625]], [[1.0]]]}, "model.layers.4.mlp.gate_up_proj": {"inputs": [[[1.8828125]]], "params": {"weight": [[1.3828125]]}}, "model.layers.4.mlp.down_proj": {"inputs": [[[0.46875]]], "outputs": [[[0.15234375]], [[0.68359375]]], "params": {"weight": [[0.73046875]]}}, "model.layers.5.self_attn.qkv_proj": {"inputs": [[[4.84375]]], "params": {"weight": [[0.42578125]]}}, "model.layers.5.self_attn.o_proj": {"inputs": [[[0.06787109375]]], "outputs": [[[0.06640625]], [[0.10888671875]]], "params": {"weight": [[0.1748046875]]}}, "model.layers.5.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.671875]], [[10.375]]]}, "model.layers.5.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.2890625]]]}, "model.layers.5.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[2.84375]]]}, "model.layers.5.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.609375]]]}, "model.layers.5.self_attn.attn.impl.k_cache": {"inputs": [[[10.375]]]}, "model.layers.5.self_attn.attn.impl.v_cache": {"inputs": [[[1.2890625]]]}, "model.layers.5.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.125]], [[9.6875]], [[1.2421875]]], "outputs": [[[0.056884765625]], [[1.0]]]}, "model.layers.5.mlp.gate_up_proj": {"inputs": [[[3.890625]]], "params": {"weight": [[10.9375]]}}, "model.layers.5.mlp.down_proj": {"inputs": [[[0.416015625]]], "outputs": [[[1.0078125]], [[7.375]]], "params": {"weight": [[0.412109375]]}}, "model.layers.6.self_attn.qkv_proj": {"inputs": [[[5.09375]]], "params": {"weight": [[0.58203125]]}}, "model.layers.6.self_attn.o_proj": {"inputs": [[[0.16796875]]], "outputs": [[[0.2451171875]], [[0.279296875]]], "params": {"weight": [[0.138671875]]}}, "model.layers.6.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.58203125]], [[16.625]]]}, "model.layers.6.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.7734375]]]}, "model.layers.6.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.40625]]]}, "model.layers.6.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.15625]]]}, "model.layers.6.self_attn.attn.impl.k_cache": {"inputs": [[[16.625]]]}, "model.layers.6.self_attn.attn.impl.v_cache": {"inputs": [[[0.7734375]]]}, "model.layers.6.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.03125]], [[15.8125]], [[0.70703125]]], "outputs": [[[0.138671875]], [[1.0]]]}, "model.layers.6.mlp.gate_up_proj": {"inputs": [[[3.328125]]], "params": {"weight": [[0.435546875]]}}, "model.layers.6.mlp.down_proj": {"inputs": [[[0.48828125]]], "outputs": [[[0.1982421875]], [[0.431640625]]], "params": {"weight": [[0.322265625]]}}, "model.layers.7.self_attn.qkv_proj": {"inputs": [[[5.625]]], "params": {"weight": [[0.5859375]]}}, "model.layers.7.self_attn.o_proj": {"inputs": [[[0.1064453125]]], "outputs": [[[0.0849609375]], [[0.1240234375]]], "params": {"weight": [[0.162109375]]}}, "model.layers.7.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.58203125]], [[14.0]]]}, "model.layers.7.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.3359375]]]}, "model.layers.7.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.9375]]]}, "model.layers.7.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.125]]]}, "model.layers.7.self_attn.attn.impl.k_cache": {"inputs": [[[14.0]]]}, "model.layers.7.self_attn.attn.impl.v_cache": {"inputs": [[[1.3359375]]]}, "model.layers.7.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.875]], [[13.1875]], [[1.234375]]], "outputs": [[[0.0703125]], [[1.0]]]}, "model.layers.7.mlp.gate_up_proj": {"inputs": [[[3.3125]]], "params": {"weight": [[0.57421875]]}}, "model.layers.7.mlp.down_proj": {"inputs": [[[0.640625]]], "outputs": [[[0.59765625]], [[1.59375]]], "params": {"weight": [[0.341796875]]}}, "model.layers.8.self_attn.qkv_proj": {"inputs": [[[4.8125]]], "params": {"weight": [[0.33984375]]}}, "model.layers.8.self_attn.o_proj": {"inputs": [[[0.1357421875]]], "outputs": [[[0.091796875]], [[0.138671875]]], "params": {"weight": [[0.10400390625]]}}, "model.layers.8.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.53125]], [[15.875]]]}, "model.layers.8.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.8671875]]]}, "model.layers.8.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[3.203125]]]}, "model.layers.8.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.15625]]]}, "model.layers.8.self_attn.attn.impl.k_cache": {"inputs": [[[15.875]]]}, "model.layers.8.self_attn.attn.impl.v_cache": {"inputs": [[[0.8671875]]]}, "model.layers.8.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.4375]], [[14.8125]], [[0.734375]]], "outputs": [[[0.0966796875]], [[1.0]]]}, "model.layers.8.mlp.gate_up_proj": {"inputs": [[[3.078125]]], "params": {"weight": [[1.8828125]]}}, "model.layers.8.mlp.down_proj": {"inputs": [[[0.8203125]]], "outputs": [[[0.126953125]], [[0.474609375]]], "params": {"weight": [[0.275390625]]}}, "model.layers.9.self_attn.qkv_proj": {"inputs": [[[5.59375]]], "params": {"weight": [[0.357421875]]}}, "model.layers.9.self_attn.o_proj": {"inputs": [[[0.2158203125]]], "outputs": [[[0.0517578125]], [[0.126953125]]], "params": {"weight": [[0.20703125]]}}, "model.layers.9.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.53515625]], [[13.625]]]}, "model.layers.9.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.0546875]]]}, "model.layers.9.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[6.25]]]}, "model.layers.9.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.0]]]}, "model.layers.9.self_attn.attn.impl.k_cache": {"inputs": [[[13.625]]]}, "model.layers.9.self_attn.attn.impl.v_cache": {"inputs": [[[1.0546875]]]}, "model.layers.9.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.25]], [[13.125]], [[0.97265625]]], "outputs": [[[0.11279296875]], [[1.0]]]}, "model.layers.9.mlp.gate_up_proj": {"inputs": [[[4.28125]]], "params": {"weight": [[18.25]]}}, "model.layers.9.mlp.down_proj": {"inputs": [[[0.58984375]]], "outputs": [[[1.4375]], [[7.15625]]], "params": {"weight": [[0.361328125]]}}, "model.layers.10.self_attn.qkv_proj": {"inputs": [[[6.4375]]], "params": {"weight": [[0.259765625]]}}, "model.layers.10.self_attn.o_proj": {"inputs": [[[0.2255859375]]], "outputs": [[[0.11181640625]], [[0.234375]]], "params": {"weight": [[0.13671875]]}}, "model.layers.10.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.64453125]], [[13.125]]]}, "model.layers.10.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.671875]]]}, "model.layers.10.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.8125]]]}, "model.layers.10.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.859375]]]}, "model.layers.10.self_attn.attn.impl.k_cache": {"inputs": [[[13.125]]]}, "model.layers.10.self_attn.attn.impl.v_cache": {"inputs": [[[1.671875]]]}, "model.layers.10.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.28125]], [[12.375]], [[1.4921875]]], "outputs": [[[0.2255859375]], [[1.0]]]}, "model.layers.10.mlp.gate_up_proj": {"inputs": [[[3.125]]], "params": {"weight": [[0.54296875]]}}, "model.layers.10.mlp.down_proj": {"inputs": [[[1.2109375]]], "outputs": [[[0.10888671875]], [[0.40625]]], "params": {"weight": [[0.373046875]]}}, "model.layers.11.self_attn.qkv_proj": {"inputs": [[[7.4375]]], "params": {"weight": [[0.52734375]]}}, "model.layers.11.self_attn.o_proj": {"inputs": [[[0.171875]]], "outputs": [[[0.08154296875]], [[0.166015625]]], "params": {"weight": [[0.232421875]]}}, "model.layers.11.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5703125]], [[12.75]]]}, "model.layers.11.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.88671875]]]}, "model.layers.11.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.1875]]]}, "model.layers.11.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.21875]]]}, "model.layers.11.self_attn.attn.impl.k_cache": {"inputs": [[[12.75]]]}, "model.layers.11.self_attn.attn.impl.v_cache": {"inputs": [[[0.88671875]]]}, "model.layers.11.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.875]], [[12.375]], [[0.79296875]]], "outputs": [[[0.150390625]], [[1.0]]]}, "model.layers.11.mlp.gate_up_proj": {"inputs": [[[3.34375]]], "params": {"weight": [[1.859375]]}}, "model.layers.11.mlp.down_proj": {"inputs": [[[0.91796875]]], "outputs": [[[0.169921875]], [[0.75390625]]], "params": {"weight": [[0.259765625]]}}, "model.layers.12.self_attn.qkv_proj": {"inputs": [[[6.1875]]], "params": {"weight": [[0.322265625]]}}, "model.layers.12.self_attn.o_proj": {"inputs": [[[0.177734375]]], "outputs": [[[0.1376953125]], [[0.32421875]]], "params": {"weight": [[0.1865234375]]}}, "model.layers.12.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.57421875]], [[14.25]]]}, "model.layers.12.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.890625]]]}, "model.layers.12.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[5.21875]]]}, "model.layers.12.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.25]]]}, "model.layers.12.self_attn.attn.impl.k_cache": {"inputs": [[[14.25]]]}, "model.layers.12.self_attn.attn.impl.v_cache": {"inputs": [[[0.890625]]]}, "model.layers.12.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.125]], [[13.4375]], [[0.796875]]], "outputs": [[[0.1728515625]], [[1.0]]]}, "model.layers.12.mlp.gate_up_proj": {"inputs": [[[3.703125]]], "params": {"weight": [[0.6875]]}}, "model.layers.12.mlp.down_proj": {"inputs": [[[0.61328125]]], "outputs": [[[0.5234375]], [[1.734375]]], "params": {"weight": [[0.2734375]]}}, "model.layers.13.self_attn.qkv_proj": {"inputs": [[[6.0]]], "params": {"weight": [[0.220703125]]}}, "model.layers.13.self_attn.o_proj": {"inputs": [[[0.291015625]]], "outputs": [[[0.150390625]], [[0.34375]]], "params": {"weight": [[0.169921875]]}}, "model.layers.13.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.64453125]], [[17.25]]]}, "model.layers.13.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.109375]]]}, "model.layers.13.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[5.15625]]]}, "model.layers.13.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.75]]]}, "model.layers.13.self_attn.attn.impl.k_cache": {"inputs": [[[17.25]]]}, "model.layers.13.self_attn.attn.impl.v_cache": {"inputs": [[[1.109375]]]}, "model.layers.13.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.96875]], [[16.875]], [[1.0234375]]], "outputs": [[[0.28125]], [[1.0]]]}, "model.layers.13.mlp.gate_up_proj": {"inputs": [[[3.8125]]], "params": {"weight": [[0.44140625]]}}, "model.layers.13.mlp.down_proj": {"inputs": [[[0.55078125]]], "outputs": [[[0.328125]], [[0.89453125]]], "params": {"weight": [[0.2392578125]]}}, "model.layers.14.self_attn.qkv_proj": {"inputs": [[[6.4375]]], "params": {"weight": [[0.283203125]]}}, "model.layers.14.self_attn.o_proj": {"inputs": [[[0.3515625]]], "outputs": [[[0.27734375]], [[0.490234375]]], "params": {"weight": [[0.13671875]]}}, "model.layers.14.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.64453125]], [[12.25]]]}, "model.layers.14.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.0703125]]]}, "model.layers.14.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[11.75]]]}, "model.layers.14.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.84375]]]}, "model.layers.14.self_attn.attn.impl.k_cache": {"inputs": [[[12.25]]]}, "model.layers.14.self_attn.attn.impl.v_cache": {"inputs": [[[1.0703125]]]}, "model.layers.14.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.15625]], [[11.3125]], [[1.03125]]], "outputs": [[[0.3125]], [[1.0]]]}, "model.layers.14.mlp.gate_up_proj": {"inputs": [[[3.875]]], "params": {"weight": [[0.60546875]]}}, "model.layers.14.mlp.down_proj": {"inputs": [[[0.73828125]]], "outputs": [[[0.306640625]], [[0.87109375]]], "params": {"weight": [[0.26953125]]}}, "model.layers.15.self_attn.qkv_proj": {"inputs": [[[7.0]]], "params": {"weight": [[0.2177734375]]}}, "model.layers.15.self_attn.o_proj": {"inputs": [[[0.302734375]]], "outputs": [[[0.1884765625]], [[0.4140625]]], "params": {"weight": [[0.162109375]]}}, "model.layers.15.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6171875]], [[10.75]]]}, "model.layers.15.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.921875]]]}, "model.layers.15.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[9.9375]]]}, "model.layers.15.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.875]]]}, "model.layers.15.self_attn.attn.impl.k_cache": {"inputs": [[[10.75]]]}, "model.layers.15.self_attn.attn.impl.v_cache": {"inputs": [[[1.921875]]]}, "model.layers.15.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.34375]], [[9.5625]], [[1.921875]]], "outputs": [[[0.302734375]], [[1.0]]]}, "model.layers.15.mlp.gate_up_proj": {"inputs": [[[3.453125]]], "params": {"weight": [[0.35546875]]}}, "model.layers.15.mlp.down_proj": {"inputs": [[[0.8203125]]], "outputs": [[[0.11328125]], [[0.375]]], "params": {"weight": [[0.2451171875]]}}, "model.layers.16.self_attn.qkv_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.2490234375]]}}, "model.layers.16.self_attn.o_proj": {"inputs": [[[0.431640625]]], "outputs": [[[0.263671875]], [[0.5234375]]], "params": {"weight": [[0.1591796875]]}}, "model.layers.16.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.72265625]], [[15.5625]]]}, "model.layers.16.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.98046875]]]}, "model.layers.16.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[9.5625]]]}, "model.layers.16.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.5625]]]}, "model.layers.16.self_attn.attn.impl.k_cache": {"inputs": [[[15.5625]]]}, "model.layers.16.self_attn.attn.impl.v_cache": {"inputs": [[[0.98046875]]]}, "model.layers.16.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.28125]], [[14.125]], [[0.85546875]]], "outputs": [[[0.326171875]], [[1.0]]]}, "model.layers.16.mlp.gate_up_proj": {"inputs": [[[4.1875]]], "params": {"weight": [[0.640625]]}}, "model.layers.16.mlp.down_proj": {"inputs": [[[1.359375]]], "outputs": [[[0.32421875]], [[0.6328125]]], "params": {"weight": [[0.283203125]]}}, "model.layers.17.self_attn.qkv_proj": {"inputs": [[[10.9375]]], "params": {"weight": [[0.28515625]]}}, "model.layers.17.self_attn.o_proj": {"inputs": [[[0.734375]]], "outputs": [[[0.236328125]], [[0.75]]], "params": {"weight": [[0.2265625]]}}, "model.layers.17.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.75]], [[12.0]]]}, "model.layers.17.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.1171875]]]}, "model.layers.17.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[23.125]]]}, "model.layers.17.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.0]]]}, "model.layers.17.self_attn.attn.impl.k_cache": {"inputs": [[[12.0]]]}, "model.layers.17.self_attn.attn.impl.v_cache": {"inputs": [[[1.1171875]]]}, "model.layers.17.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.40625]], [[11.375]], [[1.1171875]]], "outputs": [[[0.478515625]], [[1.0]]]}, "model.layers.17.mlp.gate_up_proj": {"inputs": [[[4.03125]]], "params": {"weight": [[0.388671875]]}}, "model.layers.17.mlp.down_proj": {"inputs": [[[1.578125]]], "outputs": [[[0.2353515625]], [[0.84765625]]], "params": {"weight": [[0.2265625]]}}, "model.layers.18.self_attn.qkv_proj": {"inputs": [[[11.25]]], "params": {"weight": [[0.298828125]]}}, "model.layers.18.self_attn.o_proj": {"inputs": [[[0.68359375]]], "outputs": [[[0.20703125]], [[0.640625]]], "params": {"weight": [[0.1337890625]]}}, "model.layers.18.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.68359375]], [[12.25]]]}, "model.layers.18.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.1171875]]]}, "model.layers.18.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[27.25]]]}, "model.layers.18.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.125]]]}, "model.layers.18.self_attn.attn.impl.k_cache": {"inputs": [[[12.25]]]}, "model.layers.18.self_attn.attn.impl.v_cache": {"inputs": [[[1.1171875]]]}, "model.layers.18.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.625]], [[10.375]], [[0.95703125]]], "outputs": [[[0.447265625]], [[1.0]]]}, "model.layers.18.mlp.gate_up_proj": {"inputs": [[[7.5625]]], "params": {"weight": [[0.3046875]]}}, "model.layers.18.mlp.down_proj": {"inputs": [[[2.21875]]], "outputs": [[[6.625]], [[30.625]]], "params": {"weight": [[0.81640625]]}}, "model.layers.19.self_attn.qkv_proj": {"inputs": [[[10.6875]]], "params": {"weight": [[0.2890625]]}}, "model.layers.19.self_attn.o_proj": {"inputs": [[[1.03125]]], "outputs": [[[0.28515625]], [[0.76171875]]], "params": {"weight": [[0.2177734375]]}}, "model.layers.19.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.66796875]], [[15.125]]]}, "model.layers.19.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.4609375]]]}, "model.layers.19.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[21.75]]]}, "model.layers.19.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.125]]]}, "model.layers.19.self_attn.attn.impl.k_cache": {"inputs": [[[15.125]]]}, "model.layers.19.self_attn.attn.impl.v_cache": {"inputs": [[[1.4609375]]]}, "model.layers.19.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.46875]], [[14.375]], [[1.328125]]], "outputs": [[[0.78515625]], [[1.0]]]}, "model.layers.19.mlp.gate_up_proj": {"inputs": [[[4.09375]]], "params": {"weight": [[0.28125]]}}, "model.layers.19.mlp.down_proj": {"inputs": [[[2.421875]]], "outputs": [[[0.44921875]], [[1.2265625]]], "params": {"weight": [[0.6015625]]}}, "model.layers.20.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.302734375]]}}, "model.layers.20.self_attn.o_proj": {"inputs": [[[0.66796875]]], "outputs": [[[0.2451171875]], [[1.1484375]]], "params": {"weight": [[0.162109375]]}}, "model.layers.20.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.71875]], [[15.375]]]}, "model.layers.20.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.8828125]]]}, "model.layers.20.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[29.375]]]}, "model.layers.20.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.625]]]}, "model.layers.20.self_attn.attn.impl.k_cache": {"inputs": [[[15.375]]]}, "model.layers.20.self_attn.attn.impl.v_cache": {"inputs": [[[1.8828125]]]}, "model.layers.20.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.875]], [[15.375]], [[1.8828125]]], "outputs": [[[0.609375]], [[1.0]]]}, "model.layers.20.mlp.gate_up_proj": {"inputs": [[[3.578125]]], "params": {"weight": [[0.7734375]]}}, "model.layers.20.mlp.down_proj": {"inputs": [[[2.453125]]], "outputs": [[[0.73046875]], [[4.6875]]], "params": {"weight": [[0.462890625]]}}, "model.layers.21.self_attn.qkv_proj": {"inputs": [[[11.4375]]], "params": {"weight": [[0.25390625]]}}, "model.layers.21.self_attn.o_proj": {"inputs": [[[0.435546875]]], "outputs": [[[0.2890625]], [[1.3984375]]], "params": {"weight": [[0.271484375]]}}, "model.layers.21.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.58984375]], [[11.375]]]}, "model.layers.21.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.2109375]]]}, "model.layers.21.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[9.8125]]]}, "model.layers.21.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.34375]]]}, "model.layers.21.self_attn.attn.impl.k_cache": {"inputs": [[[11.375]]]}, "model.layers.21.self_attn.attn.impl.v_cache": {"inputs": [[[1.2109375]]]}, "model.layers.21.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.59375]], [[11.25]], [[1.125]]], "outputs": [[[0.306640625]], [[1.0]]]}, "model.layers.21.mlp.gate_up_proj": {"inputs": [[[3.578125]]], "params": {"weight": [[0.3203125]]}}, "model.layers.21.mlp.down_proj": {"inputs": [[[1.8046875]]], "outputs": [[[0.68359375]], [[5.0]]], "params": {"weight": [[0.365234375]]}}, "model.layers.22.self_attn.qkv_proj": {"inputs": [[[11.0]]], "params": {"weight": [[0.3515625]]}}, "model.layers.22.self_attn.o_proj": {"inputs": [[[0.49609375]]], "outputs": [[[0.37109375]], [[1.25]]], "params": {"weight": [[0.388671875]]}}, "model.layers.22.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.81640625]], [[15.9375]]]}, "model.layers.22.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.15625]]]}, "model.layers.22.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[34.5]]]}, "model.layers.22.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[17.625]]]}, "model.layers.22.self_attn.attn.impl.k_cache": {"inputs": [[[15.9375]]]}, "model.layers.22.self_attn.attn.impl.v_cache": {"inputs": [[[1.15625]]]}, "model.layers.22.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.875]], [[15.0625]], [[1.0546875]]], "outputs": [[[0.41796875]], [[1.0]]]}, "model.layers.22.mlp.gate_up_proj": {"inputs": [[[3.25]]], "params": {"weight": [[0.34375]]}}, "model.layers.22.mlp.down_proj": {"inputs": [[[1.578125]]], "outputs": [[[0.478515625]], [[2.578125]]], "params": {"weight": [[0.41796875]]}}, "model.layers.23.self_attn.qkv_proj": {"inputs": [[[11.6875]]], "params": {"weight": [[0.40234375]]}}, "model.layers.23.self_attn.o_proj": {"inputs": [[[0.439453125]]], "outputs": [[[0.578125]], [[2.140625]]], "params": {"weight": [[0.2451171875]]}}, "model.layers.23.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.93359375]], [[15.1875]]]}, "model.layers.23.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.5546875]]]}, "model.layers.23.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[35.25]]]}, "model.layers.23.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[17.5]]]}, "model.layers.23.self_attn.attn.impl.k_cache": {"inputs": [[[15.1875]]]}, "model.layers.23.self_attn.attn.impl.v_cache": {"inputs": [[[1.5546875]]]}, "model.layers.23.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.8125]], [[13.75]], [[1.5546875]]], "outputs": [[[0.439453125]], [[1.0]]]}, "model.layers.23.mlp.gate_up_proj": {"inputs": [[[3.265625]]], "params": {"weight": [[0.33984375]]}}, "model.layers.23.mlp.down_proj": {"inputs": [[[2.359375]]], "outputs": [[[0.5625]], [[2.75]]], "params": {"weight": [[0.6015625]]}}, "model.layers.24.self_attn.qkv_proj": {"inputs": [[[9.3125]]], "params": {"weight": [[0.41796875]]}}, "model.layers.24.self_attn.o_proj": {"inputs": [[[0.75390625]]], "outputs": [[[0.2158203125]], [[2.0]]], "params": {"weight": [[0.376953125]]}}, "model.layers.24.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.015625]], [[13.75]]]}, "model.layers.24.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.296875]]]}, "model.layers.24.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[26.375]]]}, "model.layers.24.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.375]]]}, "model.layers.24.self_attn.attn.impl.k_cache": {"inputs": [[[13.75]]]}, "model.layers.24.self_attn.attn.impl.v_cache": {"inputs": [[[1.296875]]]}, "model.layers.24.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[11.125]], [[13.75]], [[1.296875]]], "outputs": [[[0.5546875]], [[1.0]]]}, "model.layers.24.mlp.gate_up_proj": {"inputs": [[[3.4375]]], "params": {"weight": [[0.69140625]]}}, "model.layers.24.mlp.down_proj": {"inputs": [[[2.5625]]], "outputs": [[[2.078125]], [[14.9375]]], "params": {"weight": [[0.5625]]}}, "model.layers.25.self_attn.qkv_proj": {"inputs": [[[10.5625]]], "params": {"weight": [[0.314453125]]}}, "model.layers.25.self_attn.o_proj": {"inputs": [[[1.0859375]]], "outputs": [[[0.375]], [[2.28125]]], "params": {"weight": [[0.44921875]]}}, "model.layers.25.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6796875]], [[11.6875]]]}, "model.layers.25.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.59375]]]}, "model.layers.25.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[39.0]]]}, "model.layers.25.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.25]]]}, "model.layers.25.self_attn.attn.impl.k_cache": {"inputs": [[[11.6875]]]}, "model.layers.25.self_attn.attn.impl.v_cache": {"inputs": [[[2.59375]]]}, "model.layers.25.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.375]], [[11.1875]], [[2.59375]]], "outputs": [[[0.365234375]], [[1.0]]]}, "model.layers.25.mlp.gate_up_proj": {"inputs": [[[3.5]]], "params": {"weight": [[0.62109375]]}}, "model.layers.25.mlp.down_proj": {"inputs": [[[3.6875]]], "outputs": [[[0.9375]], [[6.28125]]], "params": {"weight": [[1.125]]}}, "model.layers.26.self_attn.qkv_proj": {"inputs": [[[10.875]]], "params": {"weight": [[0.328125]]}}, "model.layers.26.self_attn.o_proj": {"inputs": [[[0.703125]]], "outputs": [[[0.341796875]], [[2.1875]]], "params": {"weight": [[0.3671875]]}}, "model.layers.26.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6171875]], [[14.25]]]}, "model.layers.26.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.6875]]]}, "model.layers.26.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[13.3125]]]}, "model.layers.26.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.125]]]}, "model.layers.26.self_attn.attn.impl.k_cache": {"inputs": [[[14.25]]]}, "model.layers.26.self_attn.attn.impl.v_cache": {"inputs": [[[4.6875]]]}, "model.layers.26.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.34375]], [[14.125]], [[4.6875]]], "outputs": [[[0.455078125]], [[1.0]]]}, "model.layers.26.mlp.gate_up_proj": {"inputs": [[[3.890625]]], "params": {"weight": [[0.390625]]}}, "model.layers.26.mlp.down_proj": {"inputs": [[[2.3125]]], "outputs": [[[0.984375]], [[3.546875]]], "params": {"weight": [[0.58203125]]}}, "model.layers.27.self_attn.qkv_proj": {"inputs": [[[11.0]]], "params": {"weight": [[0.265625]]}}, "model.layers.27.self_attn.o_proj": {"inputs": [[[0.62109375]]], "outputs": [[[0.9921875]], [[2.34375]]], "params": {"weight": [[0.380859375]]}}, "model.layers.27.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.69140625]], [[17.25]]]}, "model.layers.27.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.390625]]]}, "model.layers.27.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[28.625]]]}, "model.layers.27.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.5625]]]}, "model.layers.27.self_attn.attn.impl.k_cache": {"inputs": [[[17.25]]]}, "model.layers.27.self_attn.attn.impl.v_cache": {"inputs": [[[1.390625]]]}, "model.layers.27.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.9375]], [[16.5]], [[1.390625]]], "outputs": [[[0.478515625]], [[1.0]]]}, "model.layers.27.mlp.gate_up_proj": {"inputs": [[[4.0625]]], "params": {"weight": [[0.333984375]]}}, "model.layers.27.mlp.down_proj": {"inputs": [[[2.453125]]], "outputs": [[[0.85546875]], [[2.375]]], "params": {"weight": [[0.7890625]]}}, "model.layers.28.self_attn.qkv_proj": {"inputs": [[[11.0]]], "params": {"weight": [[0.32421875]]}}, "model.layers.28.self_attn.o_proj": {"inputs": [[[0.6015625]]], "outputs": [[[0.82421875]], [[2.625]]], "params": {"weight": [[0.2392578125]]}}, "model.layers.28.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.9609375]], [[19.5]]]}, "model.layers.28.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.8125]]]}, "model.layers.28.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[30.875]]]}, "model.layers.28.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.3125]]]}, "model.layers.28.self_attn.attn.impl.k_cache": {"inputs": [[[19.5]]]}, "model.layers.28.self_attn.attn.impl.v_cache": {"inputs": [[[4.8125]]]}, "model.layers.28.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.4375]], [[18.625]], [[4.8125]]], "outputs": [[[0.6015625]], [[1.0]]]}, "model.layers.28.mlp.gate_up_proj": {"inputs": [[[4.78125]]], "params": {"weight": [[0.50390625]]}}, "model.layers.28.mlp.down_proj": {"inputs": [[[2.03125]]], "outputs": [[[0.8984375]], [[2.90625]]], "params": {"weight": [[0.58203125]]}}, "model.layers.29.self_attn.qkv_proj": {"inputs": [[[11.125]]], "params": {"weight": [[0.310546875]]}}, "model.layers.29.self_attn.o_proj": {"inputs": [[[1.1328125]]], "outputs": [[[2.171875]], [[2.625]]], "params": {"weight": [[0.9921875]]}}, "model.layers.29.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.671875]], [[15.5]]]}, "model.layers.29.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[7.8125]]]}, "model.layers.29.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[48.0]]]}, "model.layers.29.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[19.75]]]}, "model.layers.29.self_attn.attn.impl.k_cache": {"inputs": [[[15.5]]]}, "model.layers.29.self_attn.attn.impl.v_cache": {"inputs": [[[7.8125]]]}, "model.layers.29.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.90625]], [[13.875]], [[7.8125]]], "outputs": [[[1.1328125]], [[1.0]]]}, "model.layers.29.mlp.gate_up_proj": {"inputs": [[[4.9375]]], "params": {"weight": [[0.79296875]]}}, "model.layers.29.mlp.down_proj": {"inputs": [[[3.4375]]], "outputs": [[[1.046875]], [[2.390625]]], "params": {"weight": [[1.484375]]}}, "model.layers.30.self_attn.qkv_proj": {"inputs": [[[10.25]]], "params": {"weight": [[0.302734375]]}}, "model.layers.30.self_attn.o_proj": {"inputs": [[[0.56640625]]], "outputs": [[[0.38671875]], [[2.03125]]], "params": {"weight": [[0.40234375]]}}, "model.layers.30.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.84375]], [[13.8125]]]}, "model.layers.30.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.234375]]]}, "model.layers.30.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[23.875]]]}, "model.layers.30.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.625]]]}, "model.layers.30.self_attn.attn.impl.k_cache": {"inputs": [[[13.8125]]]}, "model.layers.30.self_attn.attn.impl.v_cache": {"inputs": [[[2.234375]]]}, "model.layers.30.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.4375]], [[13.1875]], [[2.234375]]], "outputs": [[[0.43359375]], [[1.0]]]}, "model.layers.30.mlp.gate_up_proj": {"inputs": [[[6.5]]], "params": {"weight": [[0.341796875]]}}, "model.layers.30.mlp.down_proj": {"inputs": [[[2.484375]]], "outputs": [[[0.7890625]], [[3.375]]], "params": {"weight": [[0.6875]]}}, "model.layers.31.self_attn.qkv_proj": {"inputs": [[[12.5]]], "params": {"weight": [[0.25390625]]}}, "model.layers.31.self_attn.o_proj": {"inputs": [[[0.74609375]]], "outputs": [[[1.4921875]], [[2.265625]]], "params": {"weight": [[0.4921875]]}}, "model.layers.31.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.87109375]], [[14.5]]]}, "model.layers.31.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.890625]]]}, "model.layers.31.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[19.75]]]}, "model.layers.31.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.625]]]}, "model.layers.31.self_attn.attn.impl.k_cache": {"inputs": [[[14.5]]]}, "model.layers.31.self_attn.attn.impl.v_cache": {"inputs": [[[1.890625]]]}, "model.layers.31.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.25]], [[14.5]], [[1.890625]]], "outputs": [[[0.63671875]], [[1.0]]]}, "model.layers.31.mlp.gate_up_proj": {"inputs": [[[5.09375]]], "params": {"weight": [[0.349609375]]}}, "model.layers.31.mlp.down_proj": {"inputs": [[[2.09375]]], "outputs": [[[0.60546875]], [[1.5390625]]], "params": {"weight": [[0.609375]]}}, "model.layers.32.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.2333984375]]}}, "model.layers.32.self_attn.o_proj": {"inputs": [[[0.88671875]]], "outputs": [[[0.470703125]], [[1.5]]], "params": {"weight": [[0.23828125]]}}, "model.layers.32.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5234375]], [[13.125]]]}, "model.layers.32.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[5.3125]]]}, "model.layers.32.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[17.5]]]}, "model.layers.32.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.1875]]]}, "model.layers.32.self_attn.attn.impl.k_cache": {"inputs": [[[13.125]]]}, "model.layers.32.self_attn.attn.impl.v_cache": {"inputs": [[[5.3125]]]}, "model.layers.32.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.59375]], [[12.3125]], [[5.3125]]], "outputs": [[[0.4140625]], [[1.0]]]}, "model.layers.32.mlp.gate_up_proj": {"inputs": [[[6.65625]]], "params": {"weight": [[0.35546875]]}}, "model.layers.32.mlp.down_proj": {"inputs": [[[2.40625]]], "outputs": [[[0.703125]], [[2.703125]]], "params": {"weight": [[1.6640625]]}}, "model.layers.33.self_attn.qkv_proj": {"inputs": [[[14.25]]], "params": {"weight": [[0.3125]]}}, "model.layers.33.self_attn.o_proj": {"inputs": [[[0.71875]]], "outputs": [[[1.4921875]], [[2.65625]]], "params": {"weight": [[0.306640625]]}}, "model.layers.33.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.80078125]], [[20.375]]]}, "model.layers.33.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.03125]]]}, "model.layers.33.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[45.0]]]}, "model.layers.33.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[22.375]]]}, "model.layers.33.self_attn.attn.impl.k_cache": {"inputs": [[[20.375]]]}, "model.layers.33.self_attn.attn.impl.v_cache": {"inputs": [[[2.03125]]]}, "model.layers.33.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.0]], [[17.875]], [[2.03125]]], "outputs": [[[0.625]], [[1.0]]]}, "model.layers.33.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[125.5]]}}, "model.layers.33.mlp.down_proj": {"inputs": [[[1.859375]]], "outputs": [[[0.890625]], [[1.7578125]]], "params": {"weight": [[0.478515625]]}}, "model.layers.34.self_attn.qkv_proj": {"inputs": [[[13.25]]], "params": {"weight": [[0.8359375]]}}, "model.layers.34.self_attn.o_proj": {"inputs": [[[2.375]]], "outputs": [[[13.5625]], [[14.9375]]], "params": {"weight": [[0.84375]]}}, "model.layers.34.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.75]], [[16.75]]]}, "model.layers.34.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.6875]]]}, "model.layers.34.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[50.25]]]}, "model.layers.34.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[20.375]]]}, "model.layers.34.self_attn.attn.impl.k_cache": {"inputs": [[[16.75]]]}, "model.layers.34.self_attn.attn.impl.v_cache": {"inputs": [[[4.6875]]]}, "model.layers.34.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.5625]], [[16.75]], [[4.6875]]], "outputs": [[[2.375]], [[1.0]]]}, "model.layers.34.mlp.gate_up_proj": {"inputs": [[[7.0]]], "params": {"weight": [[0.349609375]]}}, "model.layers.34.mlp.down_proj": {"inputs": [[[3.546875]]], "outputs": [[[0.85546875]], [[3.125]]], "params": {"weight": [[0.515625]]}}, "model.layers.35.self_attn.qkv_proj": {"inputs": [[[11.8125]]], "params": {"weight": [[0.34375]]}}, "model.layers.35.self_attn.o_proj": {"inputs": [[[0.671875]]], "outputs": [[[5.90625]], [[6.71875]]], "params": {"weight": [[0.484375]]}}, "model.layers.35.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.81640625]], [[17.0]]]}, "model.layers.35.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.0]]]}, "model.layers.35.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[41.5]]]}, "model.layers.35.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[20.625]]]}, "model.layers.35.self_attn.attn.impl.k_cache": {"inputs": [[[17.0]]]}, "model.layers.35.self_attn.attn.impl.v_cache": {"inputs": [[[4.0]]]}, "model.layers.35.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.0625]], [[16.25]], [[4.0]]], "outputs": [[[0.453125]], [[1.0]]]}, "model.layers.35.mlp.gate_up_proj": {"inputs": [[[6.5]]], "params": {"weight": [[0.37109375]]}}, "model.layers.35.mlp.down_proj": {"inputs": [[[2.578125]]], "outputs": [[[0.8125]], [[4.46875]]], "params": {"weight": [[0.59375]]}}, "model.layers.36.self_attn.qkv_proj": {"inputs": [[[15.4375]]], "params": {"weight": [[0.77734375]]}}, "model.layers.36.self_attn.o_proj": {"inputs": [[[0.71484375]]], "outputs": [[[2.71875]], [[2.75]]], "params": {"weight": [[0.251953125]]}}, "model.layers.36.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.7734375]], [[13.0625]]]}, "model.layers.36.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.5]]]}, "model.layers.36.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[27.375]]]}, "model.layers.36.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.5625]]]}, "model.layers.36.self_attn.attn.impl.k_cache": {"inputs": [[[13.0625]]]}, "model.layers.36.self_attn.attn.impl.v_cache": {"inputs": [[[1.5]]]}, "model.layers.36.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.375]], [[13.0]], [[1.1328125]]], "outputs": [[[0.640625]], [[1.0]]]}, "model.layers.36.mlp.gate_up_proj": {"inputs": [[[8.125]]], "params": {"weight": [[0.498046875]]}}, "model.layers.36.mlp.down_proj": {"inputs": [[[3.8125]]], "outputs": [[[0.73828125]], [[3.40625]]], "params": {"weight": [[0.84765625]]}}, "model.layers.37.self_attn.qkv_proj": {"inputs": [[[17.625]]], "params": {"weight": [[0.62890625]]}}, "model.layers.37.self_attn.o_proj": {"inputs": [[[0.9765625]]], "outputs": [[[2.859375]], [[3.59375]]], "params": {"weight": [[0.7265625]]}}, "model.layers.37.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.7734375]], [[17.625]]]}, "model.layers.37.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.046875]]]}, "model.layers.37.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[19.875]]]}, "model.layers.37.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.1875]]]}, "model.layers.37.self_attn.attn.impl.k_cache": {"inputs": [[[17.625]]]}, "model.layers.37.self_attn.attn.impl.v_cache": {"inputs": [[[2.046875]]]}, "model.layers.37.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.9375]], [[16.125]], [[2.046875]]], "outputs": [[[0.57421875]], [[1.0]]]}, "model.layers.37.mlp.gate_up_proj": {"inputs": [[[7.4375]]], "params": {"weight": [[0.32421875]]}}, "model.layers.37.mlp.down_proj": {"inputs": [[[2.3125]]], "outputs": [[[0.78515625]], [[2.21875]]], "params": {"weight": [[0.423828125]]}}, "model.layers.38.self_attn.qkv_proj": {"inputs": [[[16.125]]], "params": {"weight": [[0.416015625]]}}, "model.layers.38.self_attn.o_proj": {"inputs": [[[0.68359375]]], "outputs": [[[0.703125]], [[1.109375]]], "params": {"weight": [[0.439453125]]}}, "model.layers.38.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6328125]], [[13.0625]]]}, "model.layers.38.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.171875]]]}, "model.layers.38.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.25]]]}, "model.layers.38.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.25]]]}, "model.layers.38.self_attn.attn.impl.k_cache": {"inputs": [[[13.0625]]]}, "model.layers.38.self_attn.attn.impl.v_cache": {"inputs": [[[1.171875]]]}, "model.layers.38.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.84375]], [[12.5]], [[1.171875]]], "outputs": [[[0.3984375]], [[1.0]]]}, "model.layers.38.mlp.gate_up_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.4921875]]}}, "model.layers.38.mlp.down_proj": {"inputs": [[[3.546875]]], "outputs": [[[2.140625]], [[3.078125]]], "params": {"weight": [[0.82421875]]}}, "model.layers.39.self_attn.qkv_proj": {"inputs": [[[17.25]]], "params": {"weight": [[0.345703125]]}}, "model.layers.39.self_attn.o_proj": {"inputs": [[[1.0390625]]], "outputs": [[[3.515625]], [[3.625]]], "params": {"weight": [[0.298828125]]}}, "model.layers.39.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.8125]], [[16.375]]]}, "model.layers.39.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.40625]]]}, "model.layers.39.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[18.75]]]}, "model.layers.39.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.5]]]}, "model.layers.39.self_attn.attn.impl.k_cache": {"inputs": [[[16.375]]]}, "model.layers.39.self_attn.attn.impl.v_cache": {"inputs": [[[1.40625]]]}, "model.layers.39.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.6875]], [[13.5]], [[1.171875]]], "outputs": [[[1.0390625]], [[1.0]]]}, "model.layers.39.mlp.gate_up_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.40234375]]}}, "model.layers.39.mlp.down_proj": {"inputs": [[[3.46875]]], "outputs": [[[1.6328125]], [[3.203125]]], "params": {"weight": [[0.6015625]]}}, "model.layers.40.self_attn.qkv_proj": {"inputs": [[[13.0]]], "params": {"weight": [[0.2578125]]}}, "model.layers.40.self_attn.o_proj": {"inputs": [[[0.69140625]]], "outputs": [[[0.9609375]], [[2.0625]]], "params": {"weight": [[0.5546875]]}}, "model.layers.40.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.74609375]], [[14.4375]]]}, "model.layers.40.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.5703125]]]}, "model.layers.40.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[11.8125]]]}, "model.layers.40.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.34375]]]}, "model.layers.40.self_attn.attn.impl.k_cache": {"inputs": [[[14.4375]]]}, "model.layers.40.self_attn.attn.impl.v_cache": {"inputs": [[[1.5703125]]]}, "model.layers.40.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.28125]], [[14.4375]], [[1.46875]]], "outputs": [[[0.48828125]], [[1.0]]]}, "model.layers.40.mlp.gate_up_proj": {"inputs": [[[7.1875]]], "params": {"weight": [[0.380859375]]}}, "model.layers.40.mlp.down_proj": {"inputs": [[[7.1875]]], "outputs": [[[0.68359375]], [[1.34375]]], "params": {"weight": [[0.890625]]}}, "model.layers.41.self_attn.qkv_proj": {"inputs": [[[16.25]]], "params": {"weight": [[0.33984375]]}}, "model.layers.41.self_attn.o_proj": {"inputs": [[[0.431640625]]], "outputs": [[[0.2255859375]], [[1.078125]]], "params": {"weight": [[0.36328125]]}}, "model.layers.41.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.55859375]], [[13.6875]]]}, "model.layers.41.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.515625]]]}, "model.layers.41.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[10.875]]]}, "model.layers.41.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.0625]]]}, "model.layers.41.self_attn.attn.impl.k_cache": {"inputs": [[[13.6875]]]}, "model.layers.41.self_attn.attn.impl.v_cache": {"inputs": [[[3.515625]]]}, "model.layers.41.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.875]], [[13.0625]], [[3.515625]]], "outputs": [[[0.431640625]], [[1.0]]]}, "model.layers.41.mlp.gate_up_proj": {"inputs": [[[7.28125]]], "params": {"weight": [[0.388671875]]}}, "model.layers.41.mlp.down_proj": {"inputs": [[[4.90625]]], "outputs": [[[0.9921875]], [[1.3984375]]], "params": {"weight": [[0.78515625]]}}, "model.layers.42.self_attn.qkv_proj": {"inputs": [[[14.8125]]], "params": {"weight": [[0.4140625]]}}, "model.layers.42.self_attn.o_proj": {"inputs": [[[0.671875]]], "outputs": [[[0.1474609375]], [[0.5]]], "params": {"weight": [[0.54296875]]}}, "model.layers.42.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.62890625]], [[12.4375]]]}, "model.layers.42.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[6.125]]]}, "model.layers.42.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[10.125]]]}, "model.layers.42.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.4375]]]}, "model.layers.42.self_attn.attn.impl.k_cache": {"inputs": [[[12.4375]]]}, "model.layers.42.self_attn.attn.impl.v_cache": {"inputs": [[[6.125]]]}, "model.layers.42.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.625]], [[11.75]], [[6.125]]], "outputs": [[[0.27734375]], [[1.0]]]}, "model.layers.42.mlp.gate_up_proj": {"inputs": [[[6.75]]], "params": {"weight": [[0.490234375]]}}, "model.layers.42.mlp.down_proj": {"inputs": [[[2.9375]]], "outputs": [[[0.85546875]], [[2.046875]]], "params": {"weight": [[1.5078125]]}}, "model.layers.43.self_attn.qkv_proj": {"inputs": [[[12.6875]]], "params": {"weight": [[0.56640625]]}}, "model.layers.43.self_attn.o_proj": {"inputs": [[[0.91015625]]], "outputs": [[[0.357421875]], [[0.734375]]], "params": {"weight": [[0.3125]]}}, "model.layers.43.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.7109375]], [[21.25]]]}, "model.layers.43.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.875]]]}, "model.layers.43.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[35.75]]]}, "model.layers.43.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.125]]]}, "model.layers.43.self_attn.attn.impl.k_cache": {"inputs": [[[21.25]]]}, "model.layers.43.self_attn.attn.impl.v_cache": {"inputs": [[[1.875]]]}, "model.layers.43.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.0]], [[18.75]], [[1.2109375]]], "outputs": [[[0.6640625]], [[1.0]]]}, "model.layers.43.mlp.gate_up_proj": {"inputs": [[[6.625]]], "params": {"weight": [[0.32421875]]}}, "model.layers.43.mlp.down_proj": {"inputs": [[[2.578125]]], "outputs": [[[0.60546875]], [[0.81640625]]], "params": {"weight": [[0.6796875]]}}, "model.layers.44.self_attn.qkv_proj": {"inputs": [[[13.625]]], "params": {"weight": [[0.3046875]]}}, "model.layers.44.self_attn.o_proj": {"inputs": [[[0.73046875]]], "outputs": [[[0.67578125]], [[0.59375]]], "params": {"weight": [[0.326171875]]}}, "model.layers.44.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6796875]], [[11.1875]]]}, "model.layers.44.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.40625]]]}, "model.layers.44.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[23.0]]]}, "model.layers.44.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[11.0625]]]}, "model.layers.44.self_attn.attn.impl.k_cache": {"inputs": [[[11.1875]]]}, "model.layers.44.self_attn.attn.impl.v_cache": {"inputs": [[[1.40625]]]}, "model.layers.44.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.0625]], [[10.375]], [[1.40625]]], "outputs": [[[0.431640625]], [[1.0]]]}, "model.layers.44.mlp.gate_up_proj": {"inputs": [[[7.09375]]], "params": {"weight": [[0.421875]]}}, "model.layers.44.mlp.down_proj": {"inputs": [[[3.359375]]], "outputs": [[[0.486328125]], [[0.984375]]], "params": {"weight": [[0.5703125]]}}, "model.layers.45.self_attn.qkv_proj": {"inputs": [[[13.75]]], "params": {"weight": [[0.408203125]]}}, "model.layers.45.self_attn.o_proj": {"inputs": [[[0.9609375]]], "outputs": [[[0.546875]], [[0.9921875]]], "params": {"weight": [[0.34765625]]}}, "model.layers.45.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.625]], [[15.0]]]}, "model.layers.45.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.3671875]]]}, "model.layers.45.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[16.25]]]}, "model.layers.45.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.15625]]]}, "model.layers.45.self_attn.attn.impl.k_cache": {"inputs": [[[15.0]]]}, "model.layers.45.self_attn.attn.impl.v_cache": {"inputs": [[[1.3671875]]]}, "model.layers.45.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.71875]], [[14.625]], [[1.1015625]]], "outputs": [[[0.56640625]], [[1.0]]]}, "model.layers.45.mlp.gate_up_proj": {"inputs": [[[7.15625]]], "params": {"weight": [[0.431640625]]}}, "model.layers.45.mlp.down_proj": {"inputs": [[[4.0625]]], "outputs": [[[1.515625]], [[2.4375]]], "params": {"weight": [[0.91015625]]}}, "model.layers.46.self_attn.qkv_proj": {"inputs": [[[12.9375]]], "params": {"weight": [[0.7265625]]}}, "model.layers.46.self_attn.o_proj": {"inputs": [[[0.283203125]]], "outputs": [[[0.14453125]], [[0.2099609375]]], "params": {"weight": [[0.18359375]]}}, "model.layers.46.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.71875]], [[16.5]]]}, "model.layers.46.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.53515625]]]}, "model.layers.46.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[13.0625]]]}, "model.layers.46.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.25]]]}, "model.layers.46.self_attn.attn.impl.k_cache": {"inputs": [[[16.5]]]}, "model.layers.46.self_attn.attn.impl.v_cache": {"inputs": [[[0.53515625]]]}, "model.layers.46.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.4375]], [[16.5]], [[0.53515625]]], "outputs": [[[0.283203125]], [[1.0]]]}, "model.layers.46.mlp.gate_up_proj": {"inputs": [[[6.4375]]], "params": {"weight": [[0.498046875]]}}, "model.layers.46.mlp.down_proj": {"inputs": [[[3.4375]]], "outputs": [[[0.359375]], [[0.671875]]], "params": {"weight": [[0.314453125]]}}, "model.layers.47.self_attn.qkv_proj": {"inputs": [[[11.9375]]], "params": {"weight": [[0.5234375]]}}, "model.layers.47.self_attn.o_proj": {"inputs": [[[0.4765625]]], "outputs": [[[0.625]], [[0.64453125]]], "params": {"weight": [[0.1767578125]]}}, "model.layers.47.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.58984375]], [[12.4375]]]}, "model.layers.47.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.91796875]]]}, "model.layers.47.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[10.875]]]}, "model.layers.47.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.0]]]}, "model.layers.47.self_attn.attn.impl.k_cache": {"inputs": [[[12.4375]]]}, "model.layers.47.self_attn.attn.impl.v_cache": {"inputs": [[[0.91796875]]]}, "model.layers.47.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.71875]], [[11.0]], [[0.80859375]]], "outputs": [[[0.455078125]], [[1.0]]]}, "model.layers.47.mlp.gate_up_proj": {"inputs": [[[6.5625]]], "params": {"weight": [[0.4296875]]}}, "model.layers.47.mlp.down_proj": {"inputs": [[[11.0]]], "outputs": [[[0.5234375]], [[1.21875]]], "params": {"weight": [[0.318359375]]}}, "model.layers.48.self_attn.qkv_proj": {"inputs": [[[11.125]]], "params": {"weight": [[0.4375]]}}, "model.layers.48.self_attn.o_proj": {"inputs": [[[1.0390625]]], "outputs": [[[0.3984375]], [[0.515625]]], "params": {"weight": [[0.44921875]]}}, "model.layers.48.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.435546875]], [[14.0625]]]}, "model.layers.48.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[6.34375]]]}, "model.layers.48.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.125]]]}, "model.layers.48.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.953125]]]}, "model.layers.48.self_attn.attn.impl.k_cache": {"inputs": [[[14.0625]]]}, "model.layers.48.self_attn.attn.impl.v_cache": {"inputs": [[[6.34375]]]}, "model.layers.48.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.5]], [[14.0625]], [[6.34375]]], "outputs": [[[1.0390625]], [[1.0]]]}, "model.layers.48.mlp.gate_up_proj": {"inputs": [[[6.84375]]], "params": {"weight": [[0.486328125]]}}, "model.layers.48.mlp.down_proj": {"inputs": [[[5.96875]]], "outputs": [[[0.447265625]], [[2.171875]]], "params": {"weight": [[0.37890625]]}}, "model.layers.49.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.458984375]]}}, "model.layers.49.self_attn.o_proj": {"inputs": [[[0.77734375]]], "outputs": [[[0.333984375]], [[0.4140625]]], "params": {"weight": [[0.36328125]]}}, "model.layers.49.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.671875]], [[16.5]]]}, "model.layers.49.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.5]]]}, "model.layers.49.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[27.75]]]}, "model.layers.49.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.0]]]}, "model.layers.49.self_attn.attn.impl.k_cache": {"inputs": [[[16.5]]]}, "model.layers.49.self_attn.attn.impl.v_cache": {"inputs": [[[1.5]]]}, "model.layers.49.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.78125]], [[15.875]], [[1.1640625]]], "outputs": [[[0.48046875]], [[1.0]]]}, "model.layers.49.mlp.gate_up_proj": {"inputs": [[[6.6875]]], "params": {"weight": [[0.58203125]]}}, "model.layers.49.mlp.down_proj": {"inputs": [[[4.21875]]], "outputs": [[[1.15625]], [[3.046875]]], "params": {"weight": [[1.3046875]]}}, "model.layers.50.self_attn.qkv_proj": {"inputs": [[[11.125]]], "params": {"weight": [[0.54296875]]}}, "model.layers.50.self_attn.o_proj": {"inputs": [[[0.455078125]]], "outputs": [[[0.2080078125]], [[0.4453125]]], "params": {"weight": [[0.205078125]]}}, "model.layers.50.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.486328125]], [[13.8125]]]}, "model.layers.50.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.71875]]]}, "model.layers.50.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[11.3125]]]}, "model.layers.50.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.3125]]]}, "model.layers.50.self_attn.attn.impl.k_cache": {"inputs": [[[13.8125]]]}, "model.layers.50.self_attn.attn.impl.v_cache": {"inputs": [[[0.71875]]]}, "model.layers.50.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.8125]], [[13.8125]], [[0.65234375]]], "outputs": [[[0.353515625]], [[1.0]]]}, "model.layers.50.mlp.gate_up_proj": {"inputs": [[[6.53125]]], "params": {"weight": [[0.486328125]]}}, "model.layers.50.mlp.down_proj": {"inputs": [[[3.359375]]], "outputs": [[[0.92578125]], [[0.8125]]], "params": {"weight": [[0.244140625]]}}, "model.layers.51.self_attn.qkv_proj": {"inputs": [[[10.875]]], "params": {"weight": [[0.392578125]]}}, "model.layers.51.self_attn.o_proj": {"inputs": [[[0.94140625]]], "outputs": [[[0.396484375]], [[0.578125]]], "params": {"weight": [[0.1689453125]]}}, "model.layers.51.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.74609375]], [[15.3125]]]}, "model.layers.51.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.25]]]}, "model.layers.51.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[14.4375]]]}, "model.layers.51.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.59375]]]}, "model.layers.51.self_attn.attn.impl.k_cache": {"inputs": [[[15.3125]]]}, "model.layers.51.self_attn.attn.impl.v_cache": {"inputs": [[[1.25]]]}, "model.layers.51.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.375]], [[12.5625]], [[1.3046875]]], "outputs": [[[0.94140625]], [[1.0]]]}, "model.layers.51.mlp.gate_up_proj": {"inputs": [[[6.46875]]], "params": {"weight": [[0.5703125]]}}, "model.layers.51.mlp.down_proj": {"inputs": [[[9.75]]], "outputs": [[[0.44921875]], [[2.796875]]], "params": {"weight": [[0.404296875]]}}, "model.layers.52.self_attn.qkv_proj": {"inputs": [[[12.375]]], "params": {"weight": [[0.392578125]]}}, "model.layers.52.self_attn.o_proj": {"inputs": [[[1.2265625]]], "outputs": [[[0.671875]], [[0.9375]]], "params": {"weight": [[0.115234375]]}}, "model.layers.52.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.7578125]], [[14.1875]]]}, "model.layers.52.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.125]]]}, "model.layers.52.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[18.625]]]}, "model.layers.52.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.1875]]]}, "model.layers.52.self_attn.attn.impl.k_cache": {"inputs": [[[14.1875]]]}, "model.layers.52.self_attn.attn.impl.v_cache": {"inputs": [[[2.125]]]}, "model.layers.52.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.46875]], [[13.125]], [[1.828125]]], "outputs": [[[0.625]], [[1.0]]]}, "model.layers.52.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[0.484375]]}}, "model.layers.52.mlp.down_proj": {"inputs": [[[4.625]]], "outputs": [[[0.85546875]], [[1.9453125]]], "params": {"weight": [[0.59375]]}}, "model.layers.53.self_attn.qkv_proj": {"inputs": [[[11.25]]], "params": {"weight": [[0.51953125]]}}, "model.layers.53.self_attn.o_proj": {"inputs": [[[0.9296875]]], "outputs": [[[0.279296875]], [[0.63671875]]], "params": {"weight": [[0.0986328125]]}}, "model.layers.53.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.97265625]], [[16.125]]]}, "model.layers.53.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.53125]]]}, "model.layers.53.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.3125]]]}, "model.layers.53.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.1875]]]}, "model.layers.53.self_attn.attn.impl.k_cache": {"inputs": [[[16.125]]]}, "model.layers.53.self_attn.attn.impl.v_cache": {"inputs": [[[1.53125]]]}, "model.layers.53.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.5]], [[16.125]], [[1.359375]]], "outputs": [[[0.5546875]], [[1.0]]]}, "model.layers.53.mlp.gate_up_proj": {"inputs": [[[6.6875]]], "params": {"weight": [[0.94140625]]}}, "model.layers.53.mlp.down_proj": {"inputs": [[[4.75]]], "outputs": [[[0.59765625]], [[1.921875]]], "params": {"weight": [[0.392578125]]}}, "model.layers.54.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.53125]]}}, "model.layers.54.self_attn.o_proj": {"inputs": [[[0.89453125]]], "outputs": [[[0.66796875]], [[1.078125]]], "params": {"weight": [[0.353515625]]}}, "model.layers.54.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5546875]], [[12.125]]]}, "model.layers.54.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.328125]]]}, "model.layers.54.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[29.0]]]}, "model.layers.54.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.8125]]]}, "model.layers.54.self_attn.attn.impl.k_cache": {"inputs": [[[12.125]]]}, "model.layers.54.self_attn.attn.impl.v_cache": {"inputs": [[[1.328125]]]}, "model.layers.54.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.1875]], [[11.3125]], [[1.1796875]]], "outputs": [[[0.66796875]], [[1.0]]]}, "model.layers.54.mlp.gate_up_proj": {"inputs": [[[6.5625]]], "params": {"weight": [[0.375]]}}, "model.layers.54.mlp.down_proj": {"inputs": [[[6.1875]]], "outputs": [[[0.6171875]], [[1.2734375]]], "params": {"weight": [[0.294921875]]}}, "model.layers.55.self_attn.qkv_proj": {"inputs": [[[10.3125]]], "params": {"weight": [[0.345703125]]}}, "model.layers.55.self_attn.o_proj": {"inputs": [[[0.84765625]]], "outputs": [[[0.9765625]], [[1.28125]]], "params": {"weight": [[0.30859375]]}}, "model.layers.55.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.62890625]], [[12.0]]]}, "model.layers.55.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.28125]]]}, "model.layers.55.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[13.125]]]}, "model.layers.55.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[11.1875]]]}, "model.layers.55.self_attn.attn.impl.k_cache": {"inputs": [[[12.0]]]}, "model.layers.55.self_attn.attn.impl.v_cache": {"inputs": [[[1.28125]]]}, "model.layers.55.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.75]], [[10.5]], [[1.21875]]], "outputs": [[[0.5234375]], [[1.0]]]}, "model.layers.55.mlp.gate_up_proj": {"inputs": [[[6.6875]]], "params": {"weight": [[0.5703125]]}}, "model.layers.55.mlp.down_proj": {"inputs": [[[5.03125]]], "outputs": [[[1.59375]], [[3.078125]]], "params": {"weight": [[0.5625]]}}, "model.layers.56.self_attn.qkv_proj": {"inputs": [[[13.6875]]], "params": {"weight": [[0.328125]]}}, "model.layers.56.self_attn.o_proj": {"inputs": [[[1.0625]]], "outputs": [[[0.72265625]], [[1.0546875]]], "params": {"weight": [[0.13671875]]}}, "model.layers.56.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.0703125]], [[15.3125]]]}, "model.layers.56.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.453125]]]}, "model.layers.56.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[24.5]]]}, "model.layers.56.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.5]]]}, "model.layers.56.self_attn.attn.impl.k_cache": {"inputs": [[[15.3125]]]}, "model.layers.56.self_attn.attn.impl.v_cache": {"inputs": [[[2.453125]]]}, "model.layers.56.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[10.375]], [[13.25]], [[1.8828125]]], "outputs": [[[0.6171875]], [[1.0]]]}, "model.layers.56.mlp.gate_up_proj": {"inputs": [[[6.90625]]], "params": {"weight": [[0.546875]]}}, "model.layers.56.mlp.down_proj": {"inputs": [[[8.875]]], "outputs": [[[2.109375]], [[2.296875]]], "params": {"weight": [[0.306640625]]}}, "model.layers.57.self_attn.qkv_proj": {"inputs": [[[12.1875]]], "params": {"weight": [[0.35546875]]}}, "model.layers.57.self_attn.o_proj": {"inputs": [[[1.078125]]], "outputs": [[[0.306640625]], [[0.7578125]]], "params": {"weight": [[0.376953125]]}}, "model.layers.57.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.0546875]], [[14.0]]]}, "model.layers.57.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.03125]]]}, "model.layers.57.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[8.3125]]]}, "model.layers.57.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.0]]]}, "model.layers.57.self_attn.attn.impl.k_cache": {"inputs": [[[14.0]]]}, "model.layers.57.self_attn.attn.impl.v_cache": {"inputs": [[[2.03125]]]}, "model.layers.57.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[11.9375]], [[14.0]], [[1.8515625]]], "outputs": [[[0.58203125]], [[1.0]]]}, "model.layers.57.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[0.359375]]}}, "model.layers.57.mlp.down_proj": {"inputs": [[[6.03125]]], "outputs": [[[0.66015625]], [[1.9609375]]], "params": {"weight": [[0.421875]]}}, "model.layers.58.self_attn.qkv_proj": {"inputs": [[[12.1875]]], "params": {"weight": [[0.5546875]]}}, "model.layers.58.self_attn.o_proj": {"inputs": [[[1.2421875]]], "outputs": [[[0.298828125]], [[0.66796875]]], "params": {"weight": [[0.2392578125]]}}, "model.layers.58.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.3828125]], [[13.625]]]}, "model.layers.58.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.84375]]]}, "model.layers.58.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[8.4375]]]}, "model.layers.58.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.21875]]]}, "model.layers.58.self_attn.attn.impl.k_cache": {"inputs": [[[13.625]]]}, "model.layers.58.self_attn.attn.impl.v_cache": {"inputs": [[[1.84375]]]}, "model.layers.58.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[14.625]], [[13.625]], [[1.6796875]]], "outputs": [[[0.921875]], [[1.0]]]}, "model.layers.58.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[0.248046875]]}}, "model.layers.58.mlp.down_proj": {"inputs": [[[8.375]]], "outputs": [[[0.423828125]], [[0.734375]]], "params": {"weight": [[0.55078125]]}}, "model.layers.59.self_attn.qkv_proj": {"inputs": [[[10.8125]]], "params": {"weight": [[0.66796875]]}}, "model.layers.59.self_attn.o_proj": {"inputs": [[[0.7421875]]], "outputs": [[[0.251953125]], [[1.0625]]], "params": {"weight": [[0.2578125]]}}, "model.layers.59.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.58984375]], [[14.25]]]}, "model.layers.59.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.046875]]]}, "model.layers.59.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[13.6875]]]}, "model.layers.59.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.625]]]}, "model.layers.59.self_attn.attn.impl.k_cache": {"inputs": [[[14.25]]]}, "model.layers.59.self_attn.attn.impl.v_cache": {"inputs": [[[2.046875]]]}, "model.layers.59.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.15625]], [[14.0]], [[2.046875]]], "outputs": [[[0.54296875]], [[1.0]]]}, "model.layers.59.mlp.gate_up_proj": {"inputs": [[[7.21875]]], "params": {"weight": [[0.45703125]]}}, "model.layers.59.mlp.down_proj": {"inputs": [[[8.25]]], "outputs": [[[0.578125]], [[7.1875]]], "params": {"weight": [[1.0]]}}, "model.layers.60.self_attn.qkv_proj": {"inputs": [[[13.3125]]], "params": {"weight": [[0.33984375]]}}, "model.layers.60.self_attn.o_proj": {"inputs": [[[0.55859375]]], "outputs": [[[0.484375]], [[0.58984375]]], "params": {"weight": [[0.4140625]]}}, "model.layers.60.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6796875]], [[15.6875]]]}, "model.layers.60.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.71875]]]}, "model.layers.60.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[22.125]]]}, "model.layers.60.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.6875]]]}, "model.layers.60.self_attn.attn.impl.k_cache": {"inputs": [[[15.6875]]]}, "model.layers.60.self_attn.attn.impl.v_cache": {"inputs": [[[1.71875]]]}, "model.layers.60.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.96875]], [[15.6875]], [[1.6328125]]], "outputs": [[[0.484375]], [[1.0]]]}, "model.layers.60.mlp.gate_up_proj": {"inputs": [[[7.34375]]], "params": {"weight": [[0.365234375]]}}, "model.layers.60.mlp.down_proj": {"inputs": [[[4.625]]], "outputs": [[[2.296875]], [[3.671875]]], "params": {"weight": [[0.80078125]]}}, "model.layers.61.self_attn.qkv_proj": {"inputs": [[[11.5]]], "params": {"weight": [[0.62890625]]}}, "model.layers.61.self_attn.o_proj": {"inputs": [[[1.2734375]]], "outputs": [[[0.26953125]], [[0.62109375]]], "params": {"weight": [[0.2041015625]]}}, "model.layers.61.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.3671875]], [[15.5]]]}, "model.layers.61.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.265625]]]}, "model.layers.61.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[28.5]]]}, "model.layers.61.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.75]]]}, "model.layers.61.self_attn.attn.impl.k_cache": {"inputs": [[[15.5]]]}, "model.layers.61.self_attn.attn.impl.v_cache": {"inputs": [[[3.265625]]]}, "model.layers.61.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[13.625]], [[14.4375]], [[3.265625]]], "outputs": [[[0.71484375]], [[1.0]]]}, "model.layers.61.mlp.gate_up_proj": {"inputs": [[[7.46875]]], "params": {"weight": [[0.345703125]]}}, "model.layers.61.mlp.down_proj": {"inputs": [[[9.25]]], "outputs": [[[0.7421875]], [[2.421875]]], "params": {"weight": [[0.431640625]]}}, "model.layers.62.self_attn.qkv_proj": {"inputs": [[[11.5625]]], "params": {"weight": [[0.69921875]]}}, "model.layers.62.self_attn.o_proj": {"inputs": [[[0.388671875]]], "outputs": [[[0.75]], [[0.76171875]]], "params": {"weight": [[0.5703125]]}}, "model.layers.62.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.431640625]], [[20.0]]]}, "model.layers.62.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.453125]]]}, "model.layers.62.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[9.875]]]}, "model.layers.62.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.3125]]]}, "model.layers.62.self_attn.attn.impl.k_cache": {"inputs": [[[20.0]]]}, "model.layers.62.self_attn.attn.impl.v_cache": {"inputs": [[[2.453125]]]}, "model.layers.62.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.53125]], [[20.0]], [[2.453125]]], "outputs": [[[0.388671875]], [[1.0]]]}, "model.layers.62.mlp.gate_up_proj": {"inputs": [[[7.625]]], "params": {"weight": [[0.40234375]]}}, "model.layers.62.mlp.down_proj": {"inputs": [[[8.875]]], "outputs": [[[1.0078125]], [[1.3359375]]], "params": {"weight": [[0.4296875]]}}, "model.layers.63.self_attn.qkv_proj": {"inputs": [[[10.1875]]], "params": {"weight": [[0.63671875]]}}, "model.layers.63.self_attn.o_proj": {"inputs": [[[0.392578125]]], "outputs": [[[0.259765625]], [[1.1875]]], "params": {"weight": [[0.2578125]]}}, "model.layers.63.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.486328125]], [[13.25]]]}, "model.layers.63.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.7421875]]]}, "model.layers.63.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[17.875]]]}, "model.layers.63.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[11.0625]]]}, "model.layers.63.self_attn.attn.impl.k_cache": {"inputs": [[[13.25]]]}, "model.layers.63.self_attn.attn.impl.v_cache": {"inputs": [[[1.7421875]]]}, "model.layers.63.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.3125]], [[12.625]], [[1.734375]]], "outputs": [[[0.392578125]], [[1.0]]]}, "model.layers.63.mlp.gate_up_proj": {"inputs": [[[7.9375]]], "params": {"weight": [[0.322265625]]}}, "model.layers.63.mlp.down_proj": {"inputs": [[[8.3125]]], "outputs": [[[0.73828125]], [[1.640625]]], "params": {"weight": [[0.88671875]]}}, "model.layers.64.self_attn.qkv_proj": {"inputs": [[[13.0]]], "params": {"weight": [[0.345703125]]}}, "model.layers.64.self_attn.o_proj": {"inputs": [[[0.83984375]]], "outputs": [[[0.341796875]], [[1.5]]], "params": {"weight": [[0.318359375]]}}, "model.layers.64.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.70703125]], [[17.75]]]}, "model.layers.64.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.5390625]]]}, "model.layers.64.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[9.0625]]]}, "model.layers.64.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.40625]]]}, "model.layers.64.self_attn.attn.impl.k_cache": {"inputs": [[[17.75]]]}, "model.layers.64.self_attn.attn.impl.v_cache": {"inputs": [[[1.5390625]]]}, "model.layers.64.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.875]], [[14.0625]], [[1.1171875]]], "outputs": [[[0.59375]], [[1.0]]]}, "model.layers.64.mlp.gate_up_proj": {"inputs": [[[8.3125]]], "params": {"weight": [[0.392578125]]}}, "model.layers.64.mlp.down_proj": {"inputs": [[[7.3125]]], "outputs": [[[0.578125]], [[2.578125]]], "params": {"weight": [[1.1484375]]}}, "model.layers.65.self_attn.qkv_proj": {"inputs": [[[14.6875]]], "params": {"weight": [[1.96875]]}}, "model.layers.65.self_attn.o_proj": {"inputs": [[[0.37109375]]], "outputs": [[[1.75]], [[1.6875]]], "params": {"weight": [[1.484375]]}}, "model.layers.65.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.8125]], [[10.6875]]]}, "model.layers.65.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.99609375]]]}, "model.layers.65.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[39.5]]]}, "model.layers.65.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[21.5]]]}, "model.layers.65.self_attn.attn.impl.k_cache": {"inputs": [[[10.6875]]]}, "model.layers.65.self_attn.attn.impl.v_cache": {"inputs": [[[0.99609375]]]}, "model.layers.65.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.9375]], [[10.6875]], [[0.7734375]]], "outputs": [[[0.37109375]], [[1.0]]]}, "model.layers.65.mlp.gate_up_proj": {"inputs": [[[8.125]]], "params": {"weight": [[0.47265625]]}}, "model.layers.65.mlp.down_proj": {"inputs": [[[6.53125]]], "outputs": [[[0.76171875]], [[2.015625]]], "params": {"weight": [[0.51171875]]}}, "model.layers.66.self_attn.qkv_proj": {"inputs": [[[11.375]]], "params": {"weight": [[0.5390625]]}}, "model.layers.66.self_attn.o_proj": {"inputs": [[[0.89453125]]], "outputs": [[[1.109375]], [[1.3359375]]], "params": {"weight": [[0.314453125]]}}, "model.layers.66.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.625]], [[13.5]]]}, "model.layers.66.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.2890625]]]}, "model.layers.66.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[16.625]]]}, "model.layers.66.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.0]]]}, "model.layers.66.self_attn.attn.impl.k_cache": {"inputs": [[[13.5]]]}, "model.layers.66.self_attn.attn.impl.v_cache": {"inputs": [[[1.2890625]]]}, "model.layers.66.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.125]], [[12.1875]], [[1.25]]], "outputs": [[[0.71875]], [[1.0]]]}, "model.layers.66.mlp.gate_up_proj": {"inputs": [[[7.78125]]], "params": {"weight": [[0.58203125]]}}, "model.layers.66.mlp.down_proj": {"inputs": [[[6.46875]]], "outputs": [[[2.5625]], [[3.5]]], "params": {"weight": [[0.37890625]]}}, "model.layers.67.self_attn.qkv_proj": {"inputs": [[[12.6875]]], "params": {"weight": [[0.41796875]]}}, "model.layers.67.self_attn.o_proj": {"inputs": [[[0.58984375]]], "outputs": [[[0.62890625]], [[1.453125]]], "params": {"weight": [[0.2734375]]}}, "model.layers.67.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.53515625]], [[13.875]]]}, "model.layers.67.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.2421875]]]}, "model.layers.67.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.46875]]]}, "model.layers.67.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.125]]]}, "model.layers.67.self_attn.attn.impl.k_cache": {"inputs": [[[13.875]]]}, "model.layers.67.self_attn.attn.impl.v_cache": {"inputs": [[[1.2421875]]]}, "model.layers.67.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.8125]], [[12.125]], [[1.046875]]], "outputs": [[[0.5625]], [[1.0]]]}, "model.layers.67.mlp.gate_up_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.392578125]]}}, "model.layers.67.mlp.down_proj": {"inputs": [[[8.5625]]], "outputs": [[[0.68359375]], [[1.5546875]]], "params": {"weight": [[0.359375]]}}, "model.layers.68.self_attn.qkv_proj": {"inputs": [[[13.75]]], "params": {"weight": [[0.384765625]]}}, "model.layers.68.self_attn.o_proj": {"inputs": [[[0.9921875]]], "outputs": [[[0.294921875]], [[2.09375]]], "params": {"weight": [[0.2138671875]]}}, "model.layers.68.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.65625]], [[16.5]]]}, "model.layers.68.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.5234375]]]}, "model.layers.68.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[13.5]]]}, "model.layers.68.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.1875]]]}, "model.layers.68.self_attn.attn.impl.k_cache": {"inputs": [[[16.5]]]}, "model.layers.68.self_attn.attn.impl.v_cache": {"inputs": [[[1.5234375]]]}, "model.layers.68.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.84375]], [[15.25]], [[1.2421875]]], "outputs": [[[0.765625]], [[1.0]]]}, "model.layers.68.mlp.gate_up_proj": {"inputs": [[[9.0625]]], "params": {"weight": [[0.56640625]]}}, "model.layers.68.mlp.down_proj": {"inputs": [[[7.625]]], "outputs": [[[1.6171875]], [[3.953125]]], "params": {"weight": [[0.302734375]]}}, "model.layers.69.self_attn.qkv_proj": {"inputs": [[[12.0]]], "params": {"weight": [[0.412109375]]}}, "model.layers.69.self_attn.o_proj": {"inputs": [[[0.79296875]]], "outputs": [[[0.58203125]], [[2.109375]]], "params": {"weight": [[0.05908203125]]}}, "model.layers.69.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.53125]], [[13.875]]]}, "model.layers.69.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.203125]]]}, "model.layers.69.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.0]]]}, "model.layers.69.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.9375]]]}, "model.layers.69.self_attn.attn.impl.k_cache": {"inputs": [[[13.875]]]}, "model.layers.69.self_attn.attn.impl.v_cache": {"inputs": [[[1.203125]]]}, "model.layers.69.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.40625]], [[12.75]], [[1.203125]]], "outputs": [[[0.58203125]], [[1.0]]]}, "model.layers.69.mlp.gate_up_proj": {"inputs": [[[8.625]]], "params": {"weight": [[0.419921875]]}}, "model.layers.69.mlp.down_proj": {"inputs": [[[8.875]]], "outputs": [[[0.875]], [[1.1015625]]], "params": {"weight": [[0.412109375]]}}, "model.layers.70.self_attn.qkv_proj": {"inputs": [[[12.625]]], "params": {"weight": [[0.32421875]]}}, "model.layers.70.self_attn.o_proj": {"inputs": [[[2.71875]]], "outputs": [[[0.484375]], [[2.5]]], "params": {"weight": [[0.3671875]]}}, "model.layers.70.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.7578125]], [[19.375]]]}, "model.layers.70.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.0625]]]}, "model.layers.70.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[14.0625]]]}, "model.layers.70.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.34375]]]}, "model.layers.70.self_attn.attn.impl.k_cache": {"inputs": [[[19.375]]]}, "model.layers.70.self_attn.attn.impl.v_cache": {"inputs": [[[4.0625]]]}, "model.layers.70.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.375]], [[17.875]], [[4.0625]]], "outputs": [[[2.140625]], [[1.0]]]}, "model.layers.70.mlp.gate_up_proj": {"inputs": [[[8.875]]], "params": {"weight": [[0.671875]]}}, "model.layers.70.mlp.down_proj": {"inputs": [[[6.75]]], "outputs": [[[1.625]], [[4.875]]], "params": {"weight": [[0.6484375]]}}, "model.layers.71.self_attn.qkv_proj": {"inputs": [[[13.0625]]], "params": {"weight": [[0.3515625]]}}, "model.layers.71.self_attn.o_proj": {"inputs": [[[1.5625]]], "outputs": [[[1.125]], [[5.34375]]], "params": {"weight": [[0.3046875]]}}, "model.layers.71.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.8359375]], [[18.75]]]}, "model.layers.71.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.484375]]]}, "model.layers.71.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[38.25]]]}, "model.layers.71.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[12.9375]]]}, "model.layers.71.self_attn.attn.impl.k_cache": {"inputs": [[[18.75]]]}, "model.layers.71.self_attn.attn.impl.v_cache": {"inputs": [[[2.484375]]]}, "model.layers.71.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.25]], [[15.8125]], [[2.015625]]], "outputs": [[[1.1171875]], [[1.0]]]}, "model.layers.71.mlp.gate_up_proj": {"inputs": [[[9.4375]]], "params": {"weight": [[0.40625]]}}, "model.layers.71.mlp.down_proj": {"inputs": [[[5.90625]]], "outputs": [[[1.703125]], [[3.953125]]], "params": {"weight": [[0.99609375]]}}, "model.layers.72.self_attn.qkv_proj": {"inputs": [[[11.6875]]], "params": {"weight": [[0.287109375]]}}, "model.layers.72.self_attn.o_proj": {"inputs": [[[1.0859375]]], "outputs": [[[0.81640625]], [[3.484375]]], "params": {"weight": [[0.251953125]]}}, "model.layers.72.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.78515625]], [[15.4375]]]}, "model.layers.72.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.875]]]}, "model.layers.72.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[15.625]]]}, "model.layers.72.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.0]]]}, "model.layers.72.self_attn.attn.impl.k_cache": {"inputs": [[[15.4375]]]}, "model.layers.72.self_attn.attn.impl.v_cache": {"inputs": [[[3.875]]]}, "model.layers.72.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.75]], [[13.625]], [[3.875]]], "outputs": [[[1.015625]], [[1.0]]]}, "model.layers.72.mlp.gate_up_proj": {"inputs": [[[9.75]]], "params": {"weight": [[0.3828125]]}}, "model.layers.72.mlp.down_proj": {"inputs": [[[7.28125]]], "outputs": [[[1.40625]], [[2.53125]]], "params": {"weight": [[0.9375]]}}, "model.layers.73.self_attn.qkv_proj": {"inputs": [[[12.0]]], "params": {"weight": [[0.359375]]}}, "model.layers.73.self_attn.o_proj": {"inputs": [[[1.6015625]]], "outputs": [[[2.40625]], [[3.78125]]], "params": {"weight": [[0.26953125]]}}, "model.layers.73.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.9453125]], [[16.875]]]}, "model.layers.73.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.234375]]]}, "model.layers.73.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[23.875]]]}, "model.layers.73.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.3125]]]}, "model.layers.73.self_attn.attn.impl.k_cache": {"inputs": [[[16.875]]]}, "model.layers.73.self_attn.attn.impl.v_cache": {"inputs": [[[3.234375]]]}, "model.layers.73.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[12.25]], [[15.8125]], [[2.46875]]], "outputs": [[[0.9453125]], [[1.0]]]}, "model.layers.73.mlp.gate_up_proj": {"inputs": [[[9.9375]]], "params": {"weight": [[0.5546875]]}}, "model.layers.73.mlp.down_proj": {"inputs": [[[11.0625]]], "outputs": [[[2.78125]], [[4.25]]], "params": {"weight": [[1.09375]]}}, "model.layers.74.self_attn.qkv_proj": {"inputs": [[[11.625]]], "params": {"weight": [[0.3359375]]}}, "model.layers.74.self_attn.o_proj": {"inputs": [[[1.6015625]]], "outputs": [[[3.15625]], [[5.59375]]], "params": {"weight": [[0.353515625]]}}, "model.layers.74.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.78515625]], [[20.875]]]}, "model.layers.74.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.21875]]]}, "model.layers.74.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[21.75]]]}, "model.layers.74.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.75]]]}, "model.layers.74.self_attn.attn.impl.k_cache": {"inputs": [[[20.875]]]}, "model.layers.74.self_attn.attn.impl.v_cache": {"inputs": [[[2.21875]]]}, "model.layers.74.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[10.0625]], [[17.875]], [[2.03125]]], "outputs": [[[1.078125]], [[1.0]]]}, "model.layers.74.mlp.gate_up_proj": {"inputs": [[[8.8125]]], "params": {"weight": [[0.333984375]]}}, "model.layers.74.mlp.down_proj": {"inputs": [[[13.875]]], "outputs": [[[2.53125]], [[6.9375]]], "params": {"weight": [[0.37890625]]}}, "model.layers.75.self_attn.qkv_proj": {"inputs": [[[11.5625]]], "params": {"weight": [[0.310546875]]}}, "model.layers.75.self_attn.o_proj": {"inputs": [[[3.171875]]], "outputs": [[[1.25]], [[6.65625]]], "params": {"weight": [[0.27734375]]}}, "model.layers.75.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.9375]], [[21.25]]]}, "model.layers.75.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.1875]]]}, "model.layers.75.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[36.25]]]}, "model.layers.75.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.75]]]}, "model.layers.75.self_attn.attn.impl.k_cache": {"inputs": [[[21.25]]]}, "model.layers.75.self_attn.attn.impl.v_cache": {"inputs": [[[4.1875]]]}, "model.layers.75.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[10.875]], [[20.375]], [[3.546875]]], "outputs": [[[1.96875]], [[1.0]]]}, "model.layers.75.mlp.gate_up_proj": {"inputs": [[[8.8125]]], "params": {"weight": [[0.65234375]]}}, "model.layers.75.mlp.down_proj": {"inputs": [[[12.9375]]], "outputs": [[[9.875]], [[12.4375]]], "params": {"weight": [[1.296875]]}}, "model.layers.76.self_attn.qkv_proj": {"inputs": [[[9.0]]], "params": {"weight": [[0.404296875]]}}, "model.layers.76.self_attn.o_proj": {"inputs": [[[1.7890625]]], "outputs": [[[3.03125]], [[5.3125]]], "params": {"weight": [[0.37109375]]}}, "model.layers.76.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.5]], [[14.875]]]}, "model.layers.76.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.46875]]]}, "model.layers.76.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[61.25]]]}, "model.layers.76.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[25.625]]]}, "model.layers.76.self_attn.attn.impl.k_cache": {"inputs": [[[14.875]]]}, "model.layers.76.self_attn.attn.impl.v_cache": {"inputs": [[[2.46875]]]}, "model.layers.76.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[13.0625]], [[15.25]], [[2.0625]]], "outputs": [[[1.34375]], [[1.0]]]}, "model.layers.76.mlp.gate_up_proj": {"inputs": [[[9.375]]], "params": {"weight": [[0.5234375]]}}, "model.layers.76.mlp.down_proj": {"inputs": [[[12.375]]], "outputs": [[[2.078125]], [[4.90625]]], "params": {"weight": [[0.65234375]]}}, "model.layers.77.self_attn.qkv_proj": {"inputs": [[[9.375]]], "params": {"weight": [[0.4375]]}}, "model.layers.77.self_attn.o_proj": {"inputs": [[[2.4375]]], "outputs": [[[2.703125]], [[3.96875]]], "params": {"weight": [[0.27734375]]}}, "model.layers.77.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.77734375]], [[14.4375]]]}, "model.layers.77.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.03125]]]}, "model.layers.77.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[15.4375]]]}, "model.layers.77.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.875]]]}, "model.layers.77.self_attn.attn.impl.k_cache": {"inputs": [[[14.4375]]]}, "model.layers.77.self_attn.attn.impl.v_cache": {"inputs": [[[3.03125]]]}, "model.layers.77.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.625]], [[12.125]], [[2.984375]]], "outputs": [[[1.7578125]], [[1.0]]]}, "model.layers.77.mlp.gate_up_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.46875]]}}, "model.layers.77.mlp.down_proj": {"inputs": [[[11.0625]]], "outputs": [[[4.03125]], [[6.8125]]], "params": {"weight": [[0.64453125]]}}, "model.layers.78.self_attn.qkv_proj": {"inputs": [[[8.625]]], "params": {"weight": [[0.8125]]}}, "model.layers.78.self_attn.o_proj": {"inputs": [[[2.125]]], "outputs": [[[2.75]], [[4.03125]]], "params": {"weight": [[0.37890625]]}}, "model.layers.78.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.73828125]], [[14.75]]]}, "model.layers.78.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.90625]]]}, "model.layers.78.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[63.75]]]}, "model.layers.78.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[27.125]]]}, "model.layers.78.self_attn.attn.impl.k_cache": {"inputs": [[[14.75]]]}, "model.layers.78.self_attn.attn.impl.v_cache": {"inputs": [[[2.90625]]]}, "model.layers.78.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.625]], [[13.0]], [[2.75]]], "outputs": [[[1.875]], [[1.0]]]}, "model.layers.78.mlp.gate_up_proj": {"inputs": [[[14.0]]], "params": {"weight": [[0.80859375]]}}, "model.layers.78.mlp.down_proj": {"inputs": [[[11.0]]], "outputs": [[[86.5]], [[125.0]]], "params": {"weight": [[0.484375]]}}, "model.layers.79.self_attn.qkv_proj": {"inputs": [[[9.375]]], "params": {"weight": [[0.44140625]]}}, "model.layers.79.self_attn.o_proj": {"inputs": [[[1.296875]]], "outputs": [[[4.125]], [[5.15625]]], "params": {"weight": [[0.546875]]}}, "model.layers.79.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6640625]], [[13.625]]]}, "model.layers.79.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.8046875]]]}, "model.layers.79.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[83.5]]]}, "model.layers.79.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[43.5]]]}, "model.layers.79.self_attn.attn.impl.k_cache": {"inputs": [[[13.625]]]}, "model.layers.79.self_attn.attn.impl.v_cache": {"inputs": [[[1.8046875]]]}, "model.layers.79.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.28125]], [[12.25]], [[1.2734375]]], "outputs": [[[0.85546875]], [[1.0]]]}, "model.layers.79.mlp.gate_up_proj": {"inputs": [[[17.25]]], "params": {"weight": [[0.62109375]]}}, "model.layers.79.mlp.down_proj": {"inputs": [[[33.0]]], "outputs": [[[155.0]], [[402.0]]], "params": {"weight": [[0.8046875]]}}, "lm_head": {"inputs": [[[81.0]]], "params": {"weight": [[0.2734375]]}}}}

quant/g3/inc_output_hooks_maxabs_0_4.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d64e1115a72cd44b8a9affdb33aa9b404b98af1aafb8e82a5b98a14f5904f8c
+size 206298

quant/g3/inc_output_hooks_maxabs_0_4_mod_list.json

@@ -0,0 +1,963 @@
+[
+    "model.layers.0.self_attn.qkv_proj",
+    "model.layers.0.self_attn.o_proj",
+    "model.layers.0.self_attn.attn.impl.matmul_qk",
+    "model.layers.0.self_attn.attn.impl.softmax",
+    "model.layers.0.self_attn.attn.impl.matmul_av",
+    "model.layers.0.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.0.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.0.self_attn.attn.impl.k_cache",
+    "model.layers.0.self_attn.attn.impl.v_cache",
+    "model.layers.0.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.0.mlp.gate_up_proj",
+    "model.layers.0.mlp.down_proj",
+    "model.layers.1.self_attn.qkv_proj",
+    "model.layers.1.self_attn.o_proj",
+    "model.layers.1.self_attn.attn.impl.matmul_qk",
+    "model.layers.1.self_attn.attn.impl.softmax",
+    "model.layers.1.self_attn.attn.impl.matmul_av",
+    "model.layers.1.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.1.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.1.self_attn.attn.impl.k_cache",
+    "model.layers.1.self_attn.attn.impl.v_cache",
+    "model.layers.1.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.1.mlp.gate_up_proj",
+    "model.layers.1.mlp.down_proj",
+    "model.layers.2.self_attn.qkv_proj",
+    "model.layers.2.self_attn.o_proj",
+    "model.layers.2.self_attn.attn.impl.matmul_qk",
+    "model.layers.2.self_attn.attn.impl.softmax",
+    "model.layers.2.self_attn.attn.impl.matmul_av",
+    "model.layers.2.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.2.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.2.self_attn.attn.impl.k_cache",
+    "model.layers.2.self_attn.attn.impl.v_cache",
+    "model.layers.2.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.2.mlp.gate_up_proj",
+    "model.layers.2.mlp.down_proj",
+    "model.layers.3.self_attn.qkv_proj",
+    "model.layers.3.self_attn.o_proj",
+    "model.layers.3.self_attn.attn.impl.matmul_qk",
+    "model.layers.3.self_attn.attn.impl.softmax",
+    "model.layers.3.self_attn.attn.impl.matmul_av",
+    "model.layers.3.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.3.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.3.self_attn.attn.impl.k_cache",
+    "model.layers.3.self_attn.attn.impl.v_cache",
+    "model.layers.3.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.3.mlp.gate_up_proj",
+    "model.layers.3.mlp.down_proj",
+    "model.layers.4.self_attn.qkv_proj",
+    "model.layers.4.self_attn.o_proj",
+    "model.layers.4.self_attn.attn.impl.matmul_qk",
+    "model.layers.4.self_attn.attn.impl.softmax",
+    "model.layers.4.self_attn.attn.impl.matmul_av",
+    "model.layers.4.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.4.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.4.self_attn.attn.impl.k_cache",
+    "model.layers.4.self_attn.attn.impl.v_cache",
+    "model.layers.4.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.4.mlp.gate_up_proj",
+    "model.layers.4.mlp.down_proj",
+    "model.layers.5.self_attn.qkv_proj",
+    "model.layers.5.self_attn.o_proj",
+    "model.layers.5.self_attn.attn.impl.matmul_qk",
+    "model.layers.5.self_attn.attn.impl.softmax",
+    "model.layers.5.self_attn.attn.impl.matmul_av",
+    "model.layers.5.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.5.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.5.self_attn.attn.impl.k_cache",
+    "model.layers.5.self_attn.attn.impl.v_cache",
+    "model.layers.5.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.5.mlp.gate_up_proj",
+    "model.layers.5.mlp.down_proj",
+    "model.layers.6.self_attn.qkv_proj",
+    "model.layers.6.self_attn.o_proj",
+    "model.layers.6.self_attn.attn.impl.matmul_qk",
+    "model.layers.6.self_attn.attn.impl.softmax",
+    "model.layers.6.self_attn.attn.impl.matmul_av",
+    "model.layers.6.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.6.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.6.self_attn.attn.impl.k_cache",
+    "model.layers.6.self_attn.attn.impl.v_cache",
+    "model.layers.6.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.6.mlp.gate_up_proj",
+    "model.layers.6.mlp.down_proj",
+    "model.layers.7.self_attn.qkv_proj",
+    "model.layers.7.self_attn.o_proj",
+    "model.layers.7.self_attn.attn.impl.matmul_qk",
+    "model.layers.7.self_attn.attn.impl.softmax",
+    "model.layers.7.self_attn.attn.impl.matmul_av",
+    "model.layers.7.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.7.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.7.self_attn.attn.impl.k_cache",
+    "model.layers.7.self_attn.attn.impl.v_cache",
+    "model.layers.7.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.7.mlp.gate_up_proj",
+    "model.layers.7.mlp.down_proj",
+    "model.layers.8.self_attn.qkv_proj",
+    "model.layers.8.self_attn.o_proj",
+    "model.layers.8.self_attn.attn.impl.matmul_qk",
+    "model.layers.8.self_attn.attn.impl.softmax",
+    "model.layers.8.self_attn.attn.impl.matmul_av",
+    "model.layers.8.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.8.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.8.self_attn.attn.impl.k_cache",
+    "model.layers.8.self_attn.attn.impl.v_cache",
+    "model.layers.8.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.8.mlp.gate_up_proj",
+    "model.layers.8.mlp.down_proj",
+    "model.layers.9.self_attn.qkv_proj",
+    "model.layers.9.self_attn.o_proj",
+    "model.layers.9.self_attn.attn.impl.matmul_qk",
+    "model.layers.9.self_attn.attn.impl.softmax",
+    "model.layers.9.self_attn.attn.impl.matmul_av",
+    "model.layers.9.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.9.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.9.self_attn.attn.impl.k_cache",
+    "model.layers.9.self_attn.attn.impl.v_cache",
+    "model.layers.9.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.9.mlp.gate_up_proj",
+    "model.layers.9.mlp.down_proj",
+    "model.layers.10.self_attn.qkv_proj",
+    "model.layers.10.self_attn.o_proj",
+    "model.layers.10.self_attn.attn.impl.matmul_qk",
+    "model.layers.10.self_attn.attn.impl.softmax",
+    "model.layers.10.self_attn.attn.impl.matmul_av",
+    "model.layers.10.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.10.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.10.self_attn.attn.impl.k_cache",
+    "model.layers.10.self_attn.attn.impl.v_cache",
+    "model.layers.10.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.10.mlp.gate_up_proj",
+    "model.layers.10.mlp.down_proj",
+    "model.layers.11.self_attn.qkv_proj",
+    "model.layers.11.self_attn.o_proj",
+    "model.layers.11.self_attn.attn.impl.matmul_qk",
+    "model.layers.11.self_attn.attn.impl.softmax",
+    "model.layers.11.self_attn.attn.impl.matmul_av",
+    "model.layers.11.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.11.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.11.self_attn.attn.impl.k_cache",
+    "model.layers.11.self_attn.attn.impl.v_cache",
+    "model.layers.11.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.11.mlp.gate_up_proj",
+    "model.layers.11.mlp.down_proj",
+    "model.layers.12.self_attn.qkv_proj",
+    "model.layers.12.self_attn.o_proj",
+    "model.layers.12.self_attn.attn.impl.matmul_qk",
+    "model.layers.12.self_attn.attn.impl.softmax",
+    "model.layers.12.self_attn.attn.impl.matmul_av",
+    "model.layers.12.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.12.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.12.self_attn.attn.impl.k_cache",
+    "model.layers.12.self_attn.attn.impl.v_cache",
+    "model.layers.12.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.12.mlp.gate_up_proj",
+    "model.layers.12.mlp.down_proj",
+    "model.layers.13.self_attn.qkv_proj",
+    "model.layers.13.self_attn.o_proj",
+    "model.layers.13.self_attn.attn.impl.matmul_qk",
+    "model.layers.13.self_attn.attn.impl.softmax",
+    "model.layers.13.self_attn.attn.impl.matmul_av",
+    "model.layers.13.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.13.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.13.self_attn.attn.impl.k_cache",
+    "model.layers.13.self_attn.attn.impl.v_cache",
+    "model.layers.13.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.13.mlp.gate_up_proj",
+    "model.layers.13.mlp.down_proj",
+    "model.layers.14.self_attn.qkv_proj",
+    "model.layers.14.self_attn.o_proj",
+    "model.layers.14.self_attn.attn.impl.matmul_qk",
+    "model.layers.14.self_attn.attn.impl.softmax",
+    "model.layers.14.self_attn.attn.impl.matmul_av",
+    "model.layers.14.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.14.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.14.self_attn.attn.impl.k_cache",
+    "model.layers.14.self_attn.attn.impl.v_cache",
+    "model.layers.14.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.14.mlp.gate_up_proj",
+    "model.layers.14.mlp.down_proj",
+    "model.layers.15.self_attn.qkv_proj",
+    "model.layers.15.self_attn.o_proj",
+    "model.layers.15.self_attn.attn.impl.matmul_qk",
+    "model.layers.15.self_attn.attn.impl.softmax",
+    "model.layers.15.self_attn.attn.impl.matmul_av",
+    "model.layers.15.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.15.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.15.self_attn.attn.impl.k_cache",
+    "model.layers.15.self_attn.attn.impl.v_cache",
+    "model.layers.15.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.15.mlp.gate_up_proj",
+    "model.layers.15.mlp.down_proj",
+    "model.layers.16.self_attn.qkv_proj",
+    "model.layers.16.self_attn.o_proj",
+    "model.layers.16.self_attn.attn.impl.matmul_qk",
+    "model.layers.16.self_attn.attn.impl.softmax",
+    "model.layers.16.self_attn.attn.impl.matmul_av",
+    "model.layers.16.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.16.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.16.self_attn.attn.impl.k_cache",
+    "model.layers.16.self_attn.attn.impl.v_cache",
+    "model.layers.16.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.16.mlp.gate_up_proj",
+    "model.layers.16.mlp.down_proj",
+    "model.layers.17.self_attn.qkv_proj",
+    "model.layers.17.self_attn.o_proj",
+    "model.layers.17.self_attn.attn.impl.matmul_qk",
+    "model.layers.17.self_attn.attn.impl.softmax",
+    "model.layers.17.self_attn.attn.impl.matmul_av",
+    "model.layers.17.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.17.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.17.self_attn.attn.impl.k_cache",
+    "model.layers.17.self_attn.attn.impl.v_cache",
+    "model.layers.17.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.17.mlp.gate_up_proj",
+    "model.layers.17.mlp.down_proj",
+    "model.layers.18.self_attn.qkv_proj",
+    "model.layers.18.self_attn.o_proj",
+    "model.layers.18.self_attn.attn.impl.matmul_qk",
+    "model.layers.18.self_attn.attn.impl.softmax",
+    "model.layers.18.self_attn.attn.impl.matmul_av",
+    "model.layers.18.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.18.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.18.self_attn.attn.impl.k_cache",
+    "model.layers.18.self_attn.attn.impl.v_cache",
+    "model.layers.18.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.18.mlp.gate_up_proj",
+    "model.layers.18.mlp.down_proj",
+    "model.layers.19.self_attn.qkv_proj",
+    "model.layers.19.self_attn.o_proj",
+    "model.layers.19.self_attn.attn.impl.matmul_qk",
+    "model.layers.19.self_attn.attn.impl.softmax",
+    "model.layers.19.self_attn.attn.impl.matmul_av",
+    "model.layers.19.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.19.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.19.self_attn.attn.impl.k_cache",
+    "model.layers.19.self_attn.attn.impl.v_cache",
+    "model.layers.19.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.19.mlp.gate_up_proj",
+    "model.layers.19.mlp.down_proj",
+    "model.layers.20.self_attn.qkv_proj",
+    "model.layers.20.self_attn.o_proj",
+    "model.layers.20.self_attn.attn.impl.matmul_qk",
+    "model.layers.20.self_attn.attn.impl.softmax",
+    "model.layers.20.self_attn.attn.impl.matmul_av",
+    "model.layers.20.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.20.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.20.self_attn.attn.impl.k_cache",
+    "model.layers.20.self_attn.attn.impl.v_cache",
+    "model.layers.20.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.20.mlp.gate_up_proj",
+    "model.layers.20.mlp.down_proj",
+    "model.layers.21.self_attn.qkv_proj",
+    "model.layers.21.self_attn.o_proj",
+    "model.layers.21.self_attn.attn.impl.matmul_qk",
+    "model.layers.21.self_attn.attn.impl.softmax",
+    "model.layers.21.self_attn.attn.impl.matmul_av",
+    "model.layers.21.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.21.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.21.self_attn.attn.impl.k_cache",
+    "model.layers.21.self_attn.attn.impl.v_cache",
+    "model.layers.21.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.21.mlp.gate_up_proj",
+    "model.layers.21.mlp.down_proj",
+    "model.layers.22.self_attn.qkv_proj",
+    "model.layers.22.self_attn.o_proj",
+    "model.layers.22.self_attn.attn.impl.matmul_qk",
+    "model.layers.22.self_attn.attn.impl.softmax",
+    "model.layers.22.self_attn.attn.impl.matmul_av",
+    "model.layers.22.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.22.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.22.self_attn.attn.impl.k_cache",
+    "model.layers.22.self_attn.attn.impl.v_cache",
+    "model.layers.22.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.22.mlp.gate_up_proj",
+    "model.layers.22.mlp.down_proj",
+    "model.layers.23.self_attn.qkv_proj",
+    "model.layers.23.self_attn.o_proj",
+    "model.layers.23.self_attn.attn.impl.matmul_qk",
+    "model.layers.23.self_attn.attn.impl.softmax",
+    "model.layers.23.self_attn.attn.impl.matmul_av",
+    "model.layers.23.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.23.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.23.self_attn.attn.impl.k_cache",
+    "model.layers.23.self_attn.attn.impl.v_cache",
+    "model.layers.23.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.23.mlp.gate_up_proj",
+    "model.layers.23.mlp.down_proj",
+    "model.layers.24.self_attn.qkv_proj",
+    "model.layers.24.self_attn.o_proj",
+    "model.layers.24.self_attn.attn.impl.matmul_qk",
+    "model.layers.24.self_attn.attn.impl.softmax",
+    "model.layers.24.self_attn.attn.impl.matmul_av",
+    "model.layers.24.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.24.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.24.self_attn.attn.impl.k_cache",
+    "model.layers.24.self_attn.attn.impl.v_cache",
+    "model.layers.24.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.24.mlp.gate_up_proj",
+    "model.layers.24.mlp.down_proj",
+    "model.layers.25.self_attn.qkv_proj",
+    "model.layers.25.self_attn.o_proj",
+    "model.layers.25.self_attn.attn.impl.matmul_qk",
+    "model.layers.25.self_attn.attn.impl.softmax",
+    "model.layers.25.self_attn.attn.impl.matmul_av",
+    "model.layers.25.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.25.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.25.self_attn.attn.impl.k_cache",
+    "model.layers.25.self_attn.attn.impl.v_cache",
+    "model.layers.25.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.25.mlp.gate_up_proj",
+    "model.layers.25.mlp.down_proj",
+    "model.layers.26.self_attn.qkv_proj",
+    "model.layers.26.self_attn.o_proj",
+    "model.layers.26.self_attn.attn.impl.matmul_qk",
+    "model.layers.26.self_attn.attn.impl.softmax",
+    "model.layers.26.self_attn.attn.impl.matmul_av",
+    "model.layers.26.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.26.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.26.self_attn.attn.impl.k_cache",
+    "model.layers.26.self_attn.attn.impl.v_cache",
+    "model.layers.26.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.26.mlp.gate_up_proj",
+    "model.layers.26.mlp.down_proj",
+    "model.layers.27.self_attn.qkv_proj",
+    "model.layers.27.self_attn.o_proj",
+    "model.layers.27.self_attn.attn.impl.matmul_qk",
+    "model.layers.27.self_attn.attn.impl.softmax",
+    "model.layers.27.self_attn.attn.impl.matmul_av",
+    "model.layers.27.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.27.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.27.self_attn.attn.impl.k_cache",
+    "model.layers.27.self_attn.attn.impl.v_cache",
+    "model.layers.27.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.27.mlp.gate_up_proj",
+    "model.layers.27.mlp.down_proj",
+    "model.layers.28.self_attn.qkv_proj",
+    "model.layers.28.self_attn.o_proj",
+    "model.layers.28.self_attn.attn.impl.matmul_qk",
+    "model.layers.28.self_attn.attn.impl.softmax",
+    "model.layers.28.self_attn.attn.impl.matmul_av",
+    "model.layers.28.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.28.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.28.self_attn.attn.impl.k_cache",
+    "model.layers.28.self_attn.attn.impl.v_cache",
+    "model.layers.28.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.28.mlp.gate_up_proj",
+    "model.layers.28.mlp.down_proj",
+    "model.layers.29.self_attn.qkv_proj",
+    "model.layers.29.self_attn.o_proj",
+    "model.layers.29.self_attn.attn.impl.matmul_qk",
+    "model.layers.29.self_attn.attn.impl.softmax",
+    "model.layers.29.self_attn.attn.impl.matmul_av",
+    "model.layers.29.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.29.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.29.self_attn.attn.impl.k_cache",
+    "model.layers.29.self_attn.attn.impl.v_cache",
+    "model.layers.29.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.29.mlp.gate_up_proj",
+    "model.layers.29.mlp.down_proj",
+    "model.layers.30.self_attn.qkv_proj",
+    "model.layers.30.self_attn.o_proj",
+    "model.layers.30.self_attn.attn.impl.matmul_qk",
+    "model.layers.30.self_attn.attn.impl.softmax",
+    "model.layers.30.self_attn.attn.impl.matmul_av",
+    "model.layers.30.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.30.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.30.self_attn.attn.impl.k_cache",
+    "model.layers.30.self_attn.attn.impl.v_cache",
+    "model.layers.30.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.30.mlp.gate_up_proj",
+    "model.layers.30.mlp.down_proj",
+    "model.layers.31.self_attn.qkv_proj",
+    "model.layers.31.self_attn.o_proj",
+    "model.layers.31.self_attn.attn.impl.matmul_qk",
+    "model.layers.31.self_attn.attn.impl.softmax",
+    "model.layers.31.self_attn.attn.impl.matmul_av",
+    "model.layers.31.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.31.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.31.self_attn.attn.impl.k_cache",
+    "model.layers.31.self_attn.attn.impl.v_cache",
+    "model.layers.31.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.31.mlp.gate_up_proj",
+    "model.layers.31.mlp.down_proj",
+    "model.layers.32.self_attn.qkv_proj",
+    "model.layers.32.self_attn.o_proj",
+    "model.layers.32.self_attn.attn.impl.matmul_qk",
+    "model.layers.32.self_attn.attn.impl.softmax",
+    "model.layers.32.self_attn.attn.impl.matmul_av",
+    "model.layers.32.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.32.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.32.self_attn.attn.impl.k_cache",
+    "model.layers.32.self_attn.attn.impl.v_cache",
+    "model.layers.32.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.32.mlp.gate_up_proj",
+    "model.layers.32.mlp.down_proj",
+    "model.layers.33.self_attn.qkv_proj",
+    "model.layers.33.self_attn.o_proj",
+    "model.layers.33.self_attn.attn.impl.matmul_qk",
+    "model.layers.33.self_attn.attn.impl.softmax",
+    "model.layers.33.self_attn.attn.impl.matmul_av",
+    "model.layers.33.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.33.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.33.self_attn.attn.impl.k_cache",
+    "model.layers.33.self_attn.attn.impl.v_cache",
+    "model.layers.33.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.33.mlp.gate_up_proj",
+    "model.layers.33.mlp.down_proj",
+    "model.layers.34.self_attn.qkv_proj",
+    "model.layers.34.self_attn.o_proj",
+    "model.layers.34.self_attn.attn.impl.matmul_qk",
+    "model.layers.34.self_attn.attn.impl.softmax",
+    "model.layers.34.self_attn.attn.impl.matmul_av",
+    "model.layers.34.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.34.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.34.self_attn.attn.impl.k_cache",
+    "model.layers.34.self_attn.attn.impl.v_cache",
+    "model.layers.34.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.34.mlp.gate_up_proj",
+    "model.layers.34.mlp.down_proj",
+    "model.layers.35.self_attn.qkv_proj",
+    "model.layers.35.self_attn.o_proj",
+    "model.layers.35.self_attn.attn.impl.matmul_qk",
+    "model.layers.35.self_attn.attn.impl.softmax",
+    "model.layers.35.self_attn.attn.impl.matmul_av",
+    "model.layers.35.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.35.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.35.self_attn.attn.impl.k_cache",
+    "model.layers.35.self_attn.attn.impl.v_cache",
+    "model.layers.35.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.35.mlp.gate_up_proj",
+    "model.layers.35.mlp.down_proj",
+    "model.layers.36.self_attn.qkv_proj",
+    "model.layers.36.self_attn.o_proj",
+    "model.layers.36.self_attn.attn.impl.matmul_qk",
+    "model.layers.36.self_attn.attn.impl.softmax",
+    "model.layers.36.self_attn.attn.impl.matmul_av",
+    "model.layers.36.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.36.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.36.self_attn.attn.impl.k_cache",
+    "model.layers.36.self_attn.attn.impl.v_cache",
+    "model.layers.36.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.36.mlp.gate_up_proj",
+    "model.layers.36.mlp.down_proj",
+    "model.layers.37.self_attn.qkv_proj",
+    "model.layers.37.self_attn.o_proj",
+    "model.layers.37.self_attn.attn.impl.matmul_qk",
+    "model.layers.37.self_attn.attn.impl.softmax",
+    "model.layers.37.self_attn.attn.impl.matmul_av",
+    "model.layers.37.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.37.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.37.self_attn.attn.impl.k_cache",
+    "model.layers.37.self_attn.attn.impl.v_cache",
+    "model.layers.37.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.37.mlp.gate_up_proj",
+    "model.layers.37.mlp.down_proj",
+    "model.layers.38.self_attn.qkv_proj",
+    "model.layers.38.self_attn.o_proj",
+    "model.layers.38.self_attn.attn.impl.matmul_qk",
+    "model.layers.38.self_attn.attn.impl.softmax",
+    "model.layers.38.self_attn.attn.impl.matmul_av",
+    "model.layers.38.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.38.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.38.self_attn.attn.impl.k_cache",
+    "model.layers.38.self_attn.attn.impl.v_cache",
+    "model.layers.38.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.38.mlp.gate_up_proj",
+    "model.layers.38.mlp.down_proj",
+    "model.layers.39.self_attn.qkv_proj",
+    "model.layers.39.self_attn.o_proj",
+    "model.layers.39.self_attn.attn.impl.matmul_qk",
+    "model.layers.39.self_attn.attn.impl.softmax",
+    "model.layers.39.self_attn.attn.impl.matmul_av",
+    "model.layers.39.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.39.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.39.self_attn.attn.impl.k_cache",
+    "model.layers.39.self_attn.attn.impl.v_cache",
+    "model.layers.39.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.39.mlp.gate_up_proj",
+    "model.layers.39.mlp.down_proj",
+    "model.layers.40.self_attn.qkv_proj",
+    "model.layers.40.self_attn.o_proj",
+    "model.layers.40.self_attn.attn.impl.matmul_qk",
+    "model.layers.40.self_attn.attn.impl.softmax",
+    "model.layers.40.self_attn.attn.impl.matmul_av",
+    "model.layers.40.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.40.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.40.self_attn.attn.impl.k_cache",
+    "model.layers.40.self_attn.attn.impl.v_cache",
+    "model.layers.40.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.40.mlp.gate_up_proj",
+    "model.layers.40.mlp.down_proj",
+    "model.layers.41.self_attn.qkv_proj",
+    "model.layers.41.self_attn.o_proj",
+    "model.layers.41.self_attn.attn.impl.matmul_qk",
+    "model.layers.41.self_attn.attn.impl.softmax",
+    "model.layers.41.self_attn.attn.impl.matmul_av",
+    "model.layers.41.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.41.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.41.self_attn.attn.impl.k_cache",
+    "model.layers.41.self_attn.attn.impl.v_cache",
+    "model.layers.41.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.41.mlp.gate_up_proj",
+    "model.layers.41.mlp.down_proj",
+    "model.layers.42.self_attn.qkv_proj",
+    "model.layers.42.self_attn.o_proj",
+    "model.layers.42.self_attn.attn.impl.matmul_qk",
+    "model.layers.42.self_attn.attn.impl.softmax",
+    "model.layers.42.self_attn.attn.impl.matmul_av",
+    "model.layers.42.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.42.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.42.self_attn.attn.impl.k_cache",
+    "model.layers.42.self_attn.attn.impl.v_cache",
+    "model.layers.42.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.42.mlp.gate_up_proj",
+    "model.layers.42.mlp.down_proj",
+    "model.layers.43.self_attn.qkv_proj",
+    "model.layers.43.self_attn.o_proj",
+    "model.layers.43.self_attn.attn.impl.matmul_qk",
+    "model.layers.43.self_attn.attn.impl.softmax",
+    "model.layers.43.self_attn.attn.impl.matmul_av",
+    "model.layers.43.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.43.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.43.self_attn.attn.impl.k_cache",
+    "model.layers.43.self_attn.attn.impl.v_cache",
+    "model.layers.43.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.43.mlp.gate_up_proj",
+    "model.layers.43.mlp.down_proj",
+    "model.layers.44.self_attn.qkv_proj",
+    "model.layers.44.self_attn.o_proj",
+    "model.layers.44.self_attn.attn.impl.matmul_qk",
+    "model.layers.44.self_attn.attn.impl.softmax",
+    "model.layers.44.self_attn.attn.impl.matmul_av",
+    "model.layers.44.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.44.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.44.self_attn.attn.impl.k_cache",
+    "model.layers.44.self_attn.attn.impl.v_cache",
+    "model.layers.44.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.44.mlp.gate_up_proj",
+    "model.layers.44.mlp.down_proj",
+    "model.layers.45.self_attn.qkv_proj",
+    "model.layers.45.self_attn.o_proj",
+    "model.layers.45.self_attn.attn.impl.matmul_qk",
+    "model.layers.45.self_attn.attn.impl.softmax",
+    "model.layers.45.self_attn.attn.impl.matmul_av",
+    "model.layers.45.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.45.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.45.self_attn.attn.impl.k_cache",
+    "model.layers.45.self_attn.attn.impl.v_cache",
+    "model.layers.45.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.45.mlp.gate_up_proj",
+    "model.layers.45.mlp.down_proj",
+    "model.layers.46.self_attn.qkv_proj",
+    "model.layers.46.self_attn.o_proj",
+    "model.layers.46.self_attn.attn.impl.matmul_qk",
+    "model.layers.46.self_attn.attn.impl.softmax",
+    "model.layers.46.self_attn.attn.impl.matmul_av",
+    "model.layers.46.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.46.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.46.self_attn.attn.impl.k_cache",
+    "model.layers.46.self_attn.attn.impl.v_cache",
+    "model.layers.46.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.46.mlp.gate_up_proj",
+    "model.layers.46.mlp.down_proj",
+    "model.layers.47.self_attn.qkv_proj",
+    "model.layers.47.self_attn.o_proj",
+    "model.layers.47.self_attn.attn.impl.matmul_qk",
+    "model.layers.47.self_attn.attn.impl.softmax",
+    "model.layers.47.self_attn.attn.impl.matmul_av",
+    "model.layers.47.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.47.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.47.self_attn.attn.impl.k_cache",
+    "model.layers.47.self_attn.attn.impl.v_cache",
+    "model.layers.47.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.47.mlp.gate_up_proj",
+    "model.layers.47.mlp.down_proj",
+    "model.layers.48.self_attn.qkv_proj",
+    "model.layers.48.self_attn.o_proj",
+    "model.layers.48.self_attn.attn.impl.matmul_qk",
+    "model.layers.48.self_attn.attn.impl.softmax",
+    "model.layers.48.self_attn.attn.impl.matmul_av",
+    "model.layers.48.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.48.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.48.self_attn.attn.impl.k_cache",
+    "model.layers.48.self_attn.attn.impl.v_cache",
+    "model.layers.48.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.48.mlp.gate_up_proj",
+    "model.layers.48.mlp.down_proj",
+    "model.layers.49.self_attn.qkv_proj",
+    "model.layers.49.self_attn.o_proj",
+    "model.layers.49.self_attn.attn.impl.matmul_qk",
+    "model.layers.49.self_attn.attn.impl.softmax",
+    "model.layers.49.self_attn.attn.impl.matmul_av",
+    "model.layers.49.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.49.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.49.self_attn.attn.impl.k_cache",
+    "model.layers.49.self_attn.attn.impl.v_cache",
+    "model.layers.49.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.49.mlp.gate_up_proj",
+    "model.layers.49.mlp.down_proj",
+    "model.layers.50.self_attn.qkv_proj",
+    "model.layers.50.self_attn.o_proj",
+    "model.layers.50.self_attn.attn.impl.matmul_qk",
+    "model.layers.50.self_attn.attn.impl.softmax",
+    "model.layers.50.self_attn.attn.impl.matmul_av",
+    "model.layers.50.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.50.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.50.self_attn.attn.impl.k_cache",
+    "model.layers.50.self_attn.attn.impl.v_cache",
+    "model.layers.50.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.50.mlp.gate_up_proj",
+    "model.layers.50.mlp.down_proj",
+    "model.layers.51.self_attn.qkv_proj",
+    "model.layers.51.self_attn.o_proj",
+    "model.layers.51.self_attn.attn.impl.matmul_qk",
+    "model.layers.51.self_attn.attn.impl.softmax",
+    "model.layers.51.self_attn.attn.impl.matmul_av",
+    "model.layers.51.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.51.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.51.self_attn.attn.impl.k_cache",
+    "model.layers.51.self_attn.attn.impl.v_cache",
+    "model.layers.51.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.51.mlp.gate_up_proj",
+    "model.layers.51.mlp.down_proj",
+    "model.layers.52.self_attn.qkv_proj",
+    "model.layers.52.self_attn.o_proj",
+    "model.layers.52.self_attn.attn.impl.matmul_qk",
+    "model.layers.52.self_attn.attn.impl.softmax",
+    "model.layers.52.self_attn.attn.impl.matmul_av",
+    "model.layers.52.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.52.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.52.self_attn.attn.impl.k_cache",
+    "model.layers.52.self_attn.attn.impl.v_cache",
+    "model.layers.52.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.52.mlp.gate_up_proj",
+    "model.layers.52.mlp.down_proj",
+    "model.layers.53.self_attn.qkv_proj",
+    "model.layers.53.self_attn.o_proj",
+    "model.layers.53.self_attn.attn.impl.matmul_qk",
+    "model.layers.53.self_attn.attn.impl.softmax",
+    "model.layers.53.self_attn.attn.impl.matmul_av",
+    "model.layers.53.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.53.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.53.self_attn.attn.impl.k_cache",
+    "model.layers.53.self_attn.attn.impl.v_cache",
+    "model.layers.53.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.53.mlp.gate_up_proj",
+    "model.layers.53.mlp.down_proj",
+    "model.layers.54.self_attn.qkv_proj",
+    "model.layers.54.self_attn.o_proj",
+    "model.layers.54.self_attn.attn.impl.matmul_qk",
+    "model.layers.54.self_attn.attn.impl.softmax",
+    "model.layers.54.self_attn.attn.impl.matmul_av",
+    "model.layers.54.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.54.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.54.self_attn.attn.impl.k_cache",
+    "model.layers.54.self_attn.attn.impl.v_cache",
+    "model.layers.54.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.54.mlp.gate_up_proj",
+    "model.layers.54.mlp.down_proj",
+    "model.layers.55.self_attn.qkv_proj",
+    "model.layers.55.self_attn.o_proj",
+    "model.layers.55.self_attn.attn.impl.matmul_qk",
+    "model.layers.55.self_attn.attn.impl.softmax",
+    "model.layers.55.self_attn.attn.impl.matmul_av",
+    "model.layers.55.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.55.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.55.self_attn.attn.impl.k_cache",
+    "model.layers.55.self_attn.attn.impl.v_cache",
+    "model.layers.55.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.55.mlp.gate_up_proj",
+    "model.layers.55.mlp.down_proj",
+    "model.layers.56.self_attn.qkv_proj",
+    "model.layers.56.self_attn.o_proj",
+    "model.layers.56.self_attn.attn.impl.matmul_qk",
+    "model.layers.56.self_attn.attn.impl.softmax",
+    "model.layers.56.self_attn.attn.impl.matmul_av",
+    "model.layers.56.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.56.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.56.self_attn.attn.impl.k_cache",
+    "model.layers.56.self_attn.attn.impl.v_cache",
+    "model.layers.56.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.56.mlp.gate_up_proj",
+    "model.layers.56.mlp.down_proj",
+    "model.layers.57.self_attn.qkv_proj",
+    "model.layers.57.self_attn.o_proj",
+    "model.layers.57.self_attn.attn.impl.matmul_qk",
+    "model.layers.57.self_attn.attn.impl.softmax",
+    "model.layers.57.self_attn.attn.impl.matmul_av",
+    "model.layers.57.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.57.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.57.self_attn.attn.impl.k_cache",
+    "model.layers.57.self_attn.attn.impl.v_cache",
+    "model.layers.57.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.57.mlp.gate_up_proj",
+    "model.layers.57.mlp.down_proj",
+    "model.layers.58.self_attn.qkv_proj",
+    "model.layers.58.self_attn.o_proj",
+    "model.layers.58.self_attn.attn.impl.matmul_qk",
+    "model.layers.58.self_attn.attn.impl.softmax",
+    "model.layers.58.self_attn.attn.impl.matmul_av",
+    "model.layers.58.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.58.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.58.self_attn.attn.impl.k_cache",
+    "model.layers.58.self_attn.attn.impl.v_cache",
+    "model.layers.58.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.58.mlp.gate_up_proj",
+    "model.layers.58.mlp.down_proj",
+    "model.layers.59.self_attn.qkv_proj",
+    "model.layers.59.self_attn.o_proj",
+    "model.layers.59.self_attn.attn.impl.matmul_qk",
+    "model.layers.59.self_attn.attn.impl.softmax",
+    "model.layers.59.self_attn.attn.impl.matmul_av",
+    "model.layers.59.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.59.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.59.self_attn.attn.impl.k_cache",
+    "model.layers.59.self_attn.attn.impl.v_cache",
+    "model.layers.59.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.59.mlp.gate_up_proj",
+    "model.layers.59.mlp.down_proj",
+    "model.layers.60.self_attn.qkv_proj",
+    "model.layers.60.self_attn.o_proj",
+    "model.layers.60.self_attn.attn.impl.matmul_qk",
+    "model.layers.60.self_attn.attn.impl.softmax",
+    "model.layers.60.self_attn.attn.impl.matmul_av",
+    "model.layers.60.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.60.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.60.self_attn.attn.impl.k_cache",
+    "model.layers.60.self_attn.attn.impl.v_cache",
+    "model.layers.60.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.60.mlp.gate_up_proj",
+    "model.layers.60.mlp.down_proj",
+    "model.layers.61.self_attn.qkv_proj",
+    "model.layers.61.self_attn.o_proj",
+    "model.layers.61.self_attn.attn.impl.matmul_qk",
+    "model.layers.61.self_attn.attn.impl.softmax",
+    "model.layers.61.self_attn.attn.impl.matmul_av",
+    "model.layers.61.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.61.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.61.self_attn.attn.impl.k_cache",
+    "model.layers.61.self_attn.attn.impl.v_cache",
+    "model.layers.61.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.61.mlp.gate_up_proj",
+    "model.layers.61.mlp.down_proj",
+    "model.layers.62.self_attn.qkv_proj",
+    "model.layers.62.self_attn.o_proj",
+    "model.layers.62.self_attn.attn.impl.matmul_qk",
+    "model.layers.62.self_attn.attn.impl.softmax",
+    "model.layers.62.self_attn.attn.impl.matmul_av",
+    "model.layers.62.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.62.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.62.self_attn.attn.impl.k_cache",
+    "model.layers.62.self_attn.attn.impl.v_cache",
+    "model.layers.62.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.62.mlp.gate_up_proj",
+    "model.layers.62.mlp.down_proj",
+    "model.layers.63.self_attn.qkv_proj",
+    "model.layers.63.self_attn.o_proj",
+    "model.layers.63.self_attn.attn.impl.matmul_qk",
+    "model.layers.63.self_attn.attn.impl.softmax",
+    "model.layers.63.self_attn.attn.impl.matmul_av",
+    "model.layers.63.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.63.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.63.self_attn.attn.impl.k_cache",
+    "model.layers.63.self_attn.attn.impl.v_cache",
+    "model.layers.63.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.63.mlp.gate_up_proj",
+    "model.layers.63.mlp.down_proj",
+    "model.layers.64.self_attn.qkv_proj",
+    "model.layers.64.self_attn.o_proj",
+    "model.layers.64.self_attn.attn.impl.matmul_qk",
+    "model.layers.64.self_attn.attn.impl.softmax",
+    "model.layers.64.self_attn.attn.impl.matmul_av",
+    "model.layers.64.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.64.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.64.self_attn.attn.impl.k_cache",
+    "model.layers.64.self_attn.attn.impl.v_cache",
+    "model.layers.64.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.64.mlp.gate_up_proj",
+    "model.layers.64.mlp.down_proj",
+    "model.layers.65.self_attn.qkv_proj",
+    "model.layers.65.self_attn.o_proj",
+    "model.layers.65.self_attn.attn.impl.matmul_qk",
+    "model.layers.65.self_attn.attn.impl.softmax",
+    "model.layers.65.self_attn.attn.impl.matmul_av",
+    "model.layers.65.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.65.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.65.self_attn.attn.impl.k_cache",
+    "model.layers.65.self_attn.attn.impl.v_cache",
+    "model.layers.65.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.65.mlp.gate_up_proj",
+    "model.layers.65.mlp.down_proj",
+    "model.layers.66.self_attn.qkv_proj",
+    "model.layers.66.self_attn.o_proj",
+    "model.layers.66.self_attn.attn.impl.matmul_qk",
+    "model.layers.66.self_attn.attn.impl.softmax",
+    "model.layers.66.self_attn.attn.impl.matmul_av",
+    "model.layers.66.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.66.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.66.self_attn.attn.impl.k_cache",
+    "model.layers.66.self_attn.attn.impl.v_cache",
+    "model.layers.66.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.66.mlp.gate_up_proj",
+    "model.layers.66.mlp.down_proj",
+    "model.layers.67.self_attn.qkv_proj",
+    "model.layers.67.self_attn.o_proj",
+    "model.layers.67.self_attn.attn.impl.matmul_qk",
+    "model.layers.67.self_attn.attn.impl.softmax",
+    "model.layers.67.self_attn.attn.impl.matmul_av",
+    "model.layers.67.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.67.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.67.self_attn.attn.impl.k_cache",
+    "model.layers.67.self_attn.attn.impl.v_cache",
+    "model.layers.67.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.67.mlp.gate_up_proj",
+    "model.layers.67.mlp.down_proj",
+    "model.layers.68.self_attn.qkv_proj",
+    "model.layers.68.self_attn.o_proj",
+    "model.layers.68.self_attn.attn.impl.matmul_qk",
+    "model.layers.68.self_attn.attn.impl.softmax",
+    "model.layers.68.self_attn.attn.impl.matmul_av",
+    "model.layers.68.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.68.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.68.self_attn.attn.impl.k_cache",
+    "model.layers.68.self_attn.attn.impl.v_cache",
+    "model.layers.68.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.68.mlp.gate_up_proj",
+    "model.layers.68.mlp.down_proj",
+    "model.layers.69.self_attn.qkv_proj",
+    "model.layers.69.self_attn.o_proj",
+    "model.layers.69.self_attn.attn.impl.matmul_qk",
+    "model.layers.69.self_attn.attn.impl.softmax",
+    "model.layers.69.self_attn.attn.impl.matmul_av",
+    "model.layers.69.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.69.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.69.self_attn.attn.impl.k_cache",
+    "model.layers.69.self_attn.attn.impl.v_cache",
+    "model.layers.69.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.69.mlp.gate_up_proj",
+    "model.layers.69.mlp.down_proj",
+    "model.layers.70.self_attn.qkv_proj",
+    "model.layers.70.self_attn.o_proj",
+    "model.layers.70.self_attn.attn.impl.matmul_qk",
+    "model.layers.70.self_attn.attn.impl.softmax",
+    "model.layers.70.self_attn.attn.impl.matmul_av",
+    "model.layers.70.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.70.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.70.self_attn.attn.impl.k_cache",
+    "model.layers.70.self_attn.attn.impl.v_cache",
+    "model.layers.70.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.70.mlp.gate_up_proj",
+    "model.layers.70.mlp.down_proj",
+    "model.layers.71.self_attn.qkv_proj",
+    "model.layers.71.self_attn.o_proj",
+    "model.layers.71.self_attn.attn.impl.matmul_qk",
+    "model.layers.71.self_attn.attn.impl.softmax",
+    "model.layers.71.self_attn.attn.impl.matmul_av",
+    "model.layers.71.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.71.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.71.self_attn.attn.impl.k_cache",
+    "model.layers.71.self_attn.attn.impl.v_cache",
+    "model.layers.71.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.71.mlp.gate_up_proj",
+    "model.layers.71.mlp.down_proj",
+    "model.layers.72.self_attn.qkv_proj",
+    "model.layers.72.self_attn.o_proj",
+    "model.layers.72.self_attn.attn.impl.matmul_qk",
+    "model.layers.72.self_attn.attn.impl.softmax",
+    "model.layers.72.self_attn.attn.impl.matmul_av",
+    "model.layers.72.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.72.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.72.self_attn.attn.impl.k_cache",
+    "model.layers.72.self_attn.attn.impl.v_cache",
+    "model.layers.72.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.72.mlp.gate_up_proj",
+    "model.layers.72.mlp.down_proj",
+    "model.layers.73.self_attn.qkv_proj",
+    "model.layers.73.self_attn.o_proj",
+    "model.layers.73.self_attn.attn.impl.matmul_qk",
+    "model.layers.73.self_attn.attn.impl.softmax",
+    "model.layers.73.self_attn.attn.impl.matmul_av",
+    "model.layers.73.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.73.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.73.self_attn.attn.impl.k_cache",
+    "model.layers.73.self_attn.attn.impl.v_cache",
+    "model.layers.73.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.73.mlp.gate_up_proj",
+    "model.layers.73.mlp.down_proj",
+    "model.layers.74.self_attn.qkv_proj",
+    "model.layers.74.self_attn.o_proj",
+    "model.layers.74.self_attn.attn.impl.matmul_qk",
+    "model.layers.74.self_attn.attn.impl.softmax",
+    "model.layers.74.self_attn.attn.impl.matmul_av",
+    "model.layers.74.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.74.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.74.self_attn.attn.impl.k_cache",
+    "model.layers.74.self_attn.attn.impl.v_cache",
+    "model.layers.74.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.74.mlp.gate_up_proj",
+    "model.layers.74.mlp.down_proj",
+    "model.layers.75.self_attn.qkv_proj",
+    "model.layers.75.self_attn.o_proj",
+    "model.layers.75.self_attn.attn.impl.matmul_qk",
+    "model.layers.75.self_attn.attn.impl.softmax",
+    "model.layers.75.self_attn.attn.impl.matmul_av",
+    "model.layers.75.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.75.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.75.self_attn.attn.impl.k_cache",
+    "model.layers.75.self_attn.attn.impl.v_cache",
+    "model.layers.75.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.75.mlp.gate_up_proj",
+    "model.layers.75.mlp.down_proj",
+    "model.layers.76.self_attn.qkv_proj",
+    "model.layers.76.self_attn.o_proj",
+    "model.layers.76.self_attn.attn.impl.matmul_qk",
+    "model.layers.76.self_attn.attn.impl.softmax",
+    "model.layers.76.self_attn.attn.impl.matmul_av",
+    "model.layers.76.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.76.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.76.self_attn.attn.impl.k_cache",
+    "model.layers.76.self_attn.attn.impl.v_cache",
+    "model.layers.76.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.76.mlp.gate_up_proj",
+    "model.layers.76.mlp.down_proj",
+    "model.layers.77.self_attn.qkv_proj",
+    "model.layers.77.self_attn.o_proj",
+    "model.layers.77.self_attn.attn.impl.matmul_qk",
+    "model.layers.77.self_attn.attn.impl.softmax",
+    "model.layers.77.self_attn.attn.impl.matmul_av",
+    "model.layers.77.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.77.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.77.self_attn.attn.impl.k_cache",
+    "model.layers.77.self_attn.attn.impl.v_cache",
+    "model.layers.77.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.77.mlp.gate_up_proj",
+    "model.layers.77.mlp.down_proj",
+    "model.layers.78.self_attn.qkv_proj",
+    "model.layers.78.self_attn.o_proj",
+    "model.layers.78.self_attn.attn.impl.matmul_qk",
+    "model.layers.78.self_attn.attn.impl.softmax",
+    "model.layers.78.self_attn.attn.impl.matmul_av",
+    "model.layers.78.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.78.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.78.self_attn.attn.impl.k_cache",
+    "model.layers.78.self_attn.attn.impl.v_cache",
+    "model.layers.78.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.78.mlp.gate_up_proj",
+    "model.layers.78.mlp.down_proj",
+    "model.layers.79.self_attn.qkv_proj",
+    "model.layers.79.self_attn.o_proj",
+    "model.layers.79.self_attn.attn.impl.matmul_qk",
+    "model.layers.79.self_attn.attn.impl.softmax",
+    "model.layers.79.self_attn.attn.impl.matmul_av",
+    "model.layers.79.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.79.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.79.self_attn.attn.impl.k_cache",
+    "model.layers.79.self_attn.attn.impl.v_cache",
+    "model.layers.79.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.79.mlp.gate_up_proj",
+    "model.layers.79.mlp.down_proj",
+    "lm_head"
+]

quant/g3/inc_output_hooks_maxabs_1_4.json

@@ -0,0 +1 @@
+{"GlobalRank": null, "LocalRank": 1, "Mode": "DynamicRange", "Nodes": {"model.layers.0.self_attn.qkv_proj": {"inputs": [[[4.09375]]], "params": {"weight": [[92.0]]}}, "model.layers.0.self_attn.o_proj": {"inputs": [[[0.115234375]]], "outputs": [[[0.140625]], [[0.248046875]]], "params": {"weight": [[0.259765625]]}}, "model.layers.0.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.63671875]], [[5.53125]]]}, "model.layers.0.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.2734375]]]}, "model.layers.0.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[452.0]]]}, "model.layers.0.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[120.0]]]}, "model.layers.0.self_attn.attn.impl.k_cache": {"inputs": [[[5.53125]]]}, "model.layers.0.self_attn.attn.impl.v_cache": {"inputs": [[[0.2734375]]]}, "model.layers.0.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.8125]], [[5.46875]], [[0.2734375]]], "outputs": [[[0.1103515625]], [[1.0]]]}, "model.layers.0.mlp.gate_up_proj": {"inputs": [[[1.453125]]], "params": {"weight": [[31.625]]}}, "model.layers.0.mlp.down_proj": {"inputs": [[[22.125]]], "outputs": [[[8.625]], [[52.5]]], "params": {"weight": [[2.375]]}}, "model.layers.1.self_attn.qkv_proj": {"inputs": [[[7.40625]]], "params": {"weight": [[9.375]]}}, "model.layers.1.self_attn.o_proj": {"inputs": [[[0.072265625]]], "outputs": [[[0.65625]], [[0.765625]]], "params": {"weight": [[0.1728515625]]}}, "model.layers.1.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.95703125]], [[11.6875]]]}, "model.layers.1.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.306640625]]]}, "model.layers.1.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[3.453125]]]}, "model.layers.1.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.78125]]]}, "model.layers.1.self_attn.attn.impl.k_cache": {"inputs": [[[11.6875]]]}, "model.layers.1.self_attn.attn.impl.v_cache": {"inputs": [[[0.306640625]]]}, "model.layers.1.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[11.0]], [[11.6875]], [[0.306640625]]], "outputs": [[[0.0703125]], [[1.0]]]}, "model.layers.1.mlp.gate_up_proj": {"inputs": [[[3.09375]]], "params": {"weight": [[5.15625]]}}, "model.layers.1.mlp.down_proj": {"inputs": [[[2.5]]], "outputs": [[[2.171875]], [[5.0625]]], "params": {"weight": [[0.74609375]]}}, "model.layers.2.self_attn.qkv_proj": {"inputs": [[[6.625]]], "params": {"weight": [[4.8125]]}}, "model.layers.2.self_attn.o_proj": {"inputs": [[[0.06884765625]]], "outputs": [[[0.044189453125]], [[0.099609375]]], "params": {"weight": [[0.283203125]]}}, "model.layers.2.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.69140625]], [[13.875]]]}, "model.layers.2.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.62109375]]]}, "model.layers.2.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.125]]]}, "model.layers.2.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.203125]]]}, "model.layers.2.self_attn.attn.impl.k_cache": {"inputs": [[[13.875]]]}, "model.layers.2.self_attn.attn.impl.v_cache": {"inputs": [[[0.62109375]]]}, "model.layers.2.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.75]], [[13.4375]], [[0.546875]]], "outputs": [[[0.05126953125]], [[1.0]]]}, "model.layers.2.mlp.gate_up_proj": {"inputs": [[[2.578125]]], "params": {"weight": [[2.03125]]}}, "model.layers.2.mlp.down_proj": {"inputs": [[[0.32421875]]], "outputs": [[[0.345703125]], [[0.72265625]]], "params": {"weight": [[0.34375]]}}, "model.layers.3.self_attn.qkv_proj": {"inputs": [[[3.796875]]], "params": {"weight": [[4.65625]]}}, "model.layers.3.self_attn.o_proj": {"inputs": [[[0.10888671875]]], "outputs": [[[0.64453125]], [[0.9453125]]], "params": {"weight": [[0.1298828125]]}}, "model.layers.3.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.58984375]], [[16.25]]]}, "model.layers.3.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.625]]]}, "model.layers.3.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.96875]]]}, "model.layers.3.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.78125]]]}, "model.layers.3.self_attn.attn.impl.k_cache": {"inputs": [[[16.25]]]}, "model.layers.3.self_attn.attn.impl.v_cache": {"inputs": [[[0.625]]]}, "model.layers.3.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.28125]], [[16.125]], [[0.56640625]]], "outputs": [[[0.10888671875]], [[1.0]]]}, "model.layers.3.mlp.gate_up_proj": {"inputs": [[[22.625]]], "params": {"weight": [[8.5]]}}, "model.layers.3.mlp.down_proj": {"inputs": [[[1.75]]], "outputs": [[[4.90625]], [[452.0]]], "params": {"weight": [[1.53125]]}}, "model.layers.4.self_attn.qkv_proj": {"inputs": [[[4.875]]], "params": {"weight": [[0.55859375]]}}, "model.layers.4.self_attn.o_proj": {"inputs": [[[0.111328125]]], "outputs": [[[0.1015625]], [[0.154296875]]], "params": {"weight": [[0.7109375]]}}, "model.layers.4.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6171875]], [[20.125]]]}, "model.layers.4.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.74609375]]]}, "model.layers.4.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[3.03125]]]}, "model.layers.4.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.34375]]]}, "model.layers.4.self_attn.attn.impl.k_cache": {"inputs": [[[20.125]]]}, "model.layers.4.self_attn.attn.impl.v_cache": {"inputs": [[[0.74609375]]]}, "model.layers.4.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.125]], [[19.5]], [[0.6484375]]], "outputs": [[[0.111328125]], [[1.0]]]}, "model.layers.4.mlp.gate_up_proj": {"inputs": [[[1.8828125]]], "params": {"weight": [[6.9375]]}}, "model.layers.4.mlp.down_proj": {"inputs": [[[0.427734375]]], "outputs": [[[0.15234375]], [[0.68359375]]], "params": {"weight": [[0.734375]]}}, "model.layers.5.self_attn.qkv_proj": {"inputs": [[[4.84375]]], "params": {"weight": [[3.8125]]}}, "model.layers.5.self_attn.o_proj": {"inputs": [[[0.09521484375]]], "outputs": [[[0.06640625]], [[0.10888671875]]], "params": {"weight": [[0.2578125]]}}, "model.layers.5.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5234375]], [[16.0]]]}, "model.layers.5.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.91796875]]]}, "model.layers.5.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.375]]]}, "model.layers.5.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.0]]]}, "model.layers.5.self_attn.attn.impl.k_cache": {"inputs": [[[16.0]]]}, "model.layers.5.self_attn.attn.impl.v_cache": {"inputs": [[[0.91796875]]]}, "model.layers.5.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.53125]], [[14.8125]], [[0.89453125]]], "outputs": [[[0.08837890625]], [[1.0]]]}, "model.layers.5.mlp.gate_up_proj": {"inputs": [[[3.890625]]], "params": {"weight": [[8.3125]]}}, "model.layers.5.mlp.down_proj": {"inputs": [[[4.625]]], "outputs": [[[1.0078125]], [[7.375]]], "params": {"weight": [[1.5546875]]}}, "model.layers.6.self_attn.qkv_proj": {"inputs": [[[5.09375]]], "params": {"weight": [[0.294921875]]}}, "model.layers.6.self_attn.o_proj": {"inputs": [[[0.0888671875]]], "outputs": [[[0.2451171875]], [[0.279296875]]], "params": {"weight": [[0.12060546875]]}}, "model.layers.6.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.55859375]], [[14.125]]]}, "model.layers.6.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.078125]]]}, "model.layers.6.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[2.875]]]}, "model.layers.6.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.53125]]]}, "model.layers.6.self_attn.attn.impl.k_cache": {"inputs": [[[14.125]]]}, "model.layers.6.self_attn.attn.impl.v_cache": {"inputs": [[[1.078125]]]}, "model.layers.6.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.15625]], [[13.3125]], [[0.91796875]]], "outputs": [[[0.0888671875]], [[1.0]]]}, "model.layers.6.mlp.gate_up_proj": {"inputs": [[[3.328125]]], "params": {"weight": [[0.90625]]}}, "model.layers.6.mlp.down_proj": {"inputs": [[[0.43359375]]], "outputs": [[[0.1982421875]], [[0.431640625]]], "params": {"weight": [[0.26171875]]}}, "model.layers.7.self_attn.qkv_proj": {"inputs": [[[5.625]]], "params": {"weight": [[0.423828125]]}}, "model.layers.7.self_attn.o_proj": {"inputs": [[[0.1708984375]]], "outputs": [[[0.0849609375]], [[0.1240234375]]], "params": {"weight": [[0.181640625]]}}, "model.layers.7.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.625]], [[12.0]]]}, "model.layers.7.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.6328125]]]}, "model.layers.7.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.71875]]]}, "model.layers.7.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.59375]]]}, "model.layers.7.self_attn.attn.impl.k_cache": {"inputs": [[[12.0]]]}, "model.layers.7.self_attn.attn.impl.v_cache": {"inputs": [[[1.6328125]]]}, "model.layers.7.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.875]], [[11.125]], [[1.6328125]]], "outputs": [[[0.13671875]], [[1.0]]]}, "model.layers.7.mlp.gate_up_proj": {"inputs": [[[3.3125]]], "params": {"weight": [[0.40234375]]}}, "model.layers.7.mlp.down_proj": {"inputs": [[[0.62109375]]], "outputs": [[[0.59765625]], [[1.59375]]], "params": {"weight": [[0.248046875]]}}, "model.layers.8.self_attn.qkv_proj": {"inputs": [[[4.8125]]], "params": {"weight": [[0.5234375]]}}, "model.layers.8.self_attn.o_proj": {"inputs": [[[0.09716796875]]], "outputs": [[[0.091796875]], [[0.138671875]]], "params": {"weight": [[0.34765625]]}}, "model.layers.8.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.640625]], [[21.375]]]}, "model.layers.8.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.703125]]]}, "model.layers.8.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[2.78125]]]}, "model.layers.8.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.484375]]]}, "model.layers.8.self_attn.attn.impl.k_cache": {"inputs": [[[21.375]]]}, "model.layers.8.self_attn.attn.impl.v_cache": {"inputs": [[[1.703125]]]}, "model.layers.8.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.25]], [[20.625]], [[1.4140625]]], "outputs": [[[0.08056640625]], [[1.0]]]}, "model.layers.8.mlp.gate_up_proj": {"inputs": [[[3.078125]]], "params": {"weight": [[4.0625]]}}, "model.layers.8.mlp.down_proj": {"inputs": [[[0.55859375]]], "outputs": [[[0.126953125]], [[0.474609375]]], "params": {"weight": [[0.265625]]}}, "model.layers.9.self_attn.qkv_proj": {"inputs": [[[5.59375]]], "params": {"weight": [[1.3515625]]}}, "model.layers.9.self_attn.o_proj": {"inputs": [[[0.1259765625]]], "outputs": [[[0.0517578125]], [[0.126953125]]], "params": {"weight": [[0.16796875]]}}, "model.layers.9.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.58984375]], [[20.75]]]}, "model.layers.9.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.98046875]]]}, "model.layers.9.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[2.875]]]}, "model.layers.9.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.5625]]]}, "model.layers.9.self_attn.attn.impl.k_cache": {"inputs": [[[20.75]]]}, "model.layers.9.self_attn.attn.impl.v_cache": {"inputs": [[[0.98046875]]]}, "model.layers.9.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.0]], [[20.375]], [[0.98046875]]], "outputs": [[[0.1259765625]], [[1.0]]]}, "model.layers.9.mlp.gate_up_proj": {"inputs": [[[4.28125]]], "params": {"weight": [[5.15625]]}}, "model.layers.9.mlp.down_proj": {"inputs": [[[1.8671875]]], "outputs": [[[1.4375]], [[7.15625]]], "params": {"weight": [[0.494140625]]}}, "model.layers.10.self_attn.qkv_proj": {"inputs": [[[6.4375]]], "params": {"weight": [[0.234375]]}}, "model.layers.10.self_attn.o_proj": {"inputs": [[[0.22265625]]], "outputs": [[[0.11181640625]], [[0.234375]]], "params": {"weight": [[0.166015625]]}}, "model.layers.10.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.84375]], [[10.25]]]}, "model.layers.10.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.6875]]]}, "model.layers.10.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.9375]]]}, "model.layers.10.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.46875]]]}, "model.layers.10.self_attn.attn.impl.k_cache": {"inputs": [[[10.25]]]}, "model.layers.10.self_attn.attn.impl.v_cache": {"inputs": [[[1.6875]]]}, "model.layers.10.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.75]], [[9.375]], [[1.6875]]], "outputs": [[[0.22265625]], [[1.0]]]}, "model.layers.10.mlp.gate_up_proj": {"inputs": [[[3.125]]], "params": {"weight": [[0.302734375]]}}, "model.layers.10.mlp.down_proj": {"inputs": [[[0.71875]]], "outputs": [[[0.10888671875]], [[0.40625]]], "params": {"weight": [[0.271484375]]}}, "model.layers.11.self_attn.qkv_proj": {"inputs": [[[7.4375]]], "params": {"weight": [[0.98828125]]}}, "model.layers.11.self_attn.o_proj": {"inputs": [[[0.302734375]]], "outputs": [[[0.08154296875]], [[0.166015625]]], "params": {"weight": [[0.22265625]]}}, "model.layers.11.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.69921875]], [[17.125]]]}, "model.layers.11.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.1796875]]]}, "model.layers.11.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[8.125]]]}, "model.layers.11.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.375]]]}, "model.layers.11.self_attn.attn.impl.k_cache": {"inputs": [[[17.125]]]}, "model.layers.11.self_attn.attn.impl.v_cache": {"inputs": [[[1.1796875]]]}, "model.layers.11.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.21875]], [[16.375]], [[1.1796875]]], "outputs": [[[0.302734375]], [[1.0]]]}, "model.layers.11.mlp.gate_up_proj": {"inputs": [[[3.34375]]], "params": {"weight": [[1.4609375]]}}, "model.layers.11.mlp.down_proj": {"inputs": [[[0.73828125]]], "outputs": [[[0.169921875]], [[0.75390625]]], "params": {"weight": [[0.37890625]]}}, "model.layers.12.self_attn.qkv_proj": {"inputs": [[[6.1875]]], "params": {"weight": [[0.265625]]}}, "model.layers.12.self_attn.o_proj": {"inputs": [[[0.1572265625]]], "outputs": [[[0.1376953125]], [[0.32421875]]], "params": {"weight": [[0.19921875]]}}, "model.layers.12.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.62890625]], [[19.125]]]}, "model.layers.12.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.0859375]]]}, "model.layers.12.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[5.5]]]}, "model.layers.12.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.5]]]}, "model.layers.12.self_attn.attn.impl.k_cache": {"inputs": [[[19.125]]]}, "model.layers.12.self_attn.attn.impl.v_cache": {"inputs": [[[1.0859375]]]}, "model.layers.12.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.1875]], [[18.875]], [[1.0859375]]], "outputs": [[[0.103515625]], [[1.0]]]}, "model.layers.12.mlp.gate_up_proj": {"inputs": [[[3.703125]]], "params": {"weight": [[0.3125]]}}, "model.layers.12.mlp.down_proj": {"inputs": [[[2.15625]]], "outputs": [[[0.5234375]], [[1.734375]]], "params": {"weight": [[0.267578125]]}}, "model.layers.13.self_attn.qkv_proj": {"inputs": [[[6.0]]], "params": {"weight": [[0.267578125]]}}, "model.layers.13.self_attn.o_proj": {"inputs": [[[0.48828125]]], "outputs": [[[0.150390625]], [[0.34375]]], "params": {"weight": [[0.1943359375]]}}, "model.layers.13.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.734375]], [[13.1875]]]}, "model.layers.13.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.21875]]]}, "model.layers.13.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[9.0625]]]}, "model.layers.13.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.03125]]]}, "model.layers.13.self_attn.attn.impl.k_cache": {"inputs": [[[13.1875]]]}, "model.layers.13.self_attn.attn.impl.v_cache": {"inputs": [[[1.21875]]]}, "model.layers.13.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.71875]], [[13.0625]], [[1.21875]]], "outputs": [[[0.30078125]], [[1.0]]]}, "model.layers.13.mlp.gate_up_proj": {"inputs": [[[3.8125]]], "params": {"weight": [[0.423828125]]}}, "model.layers.13.mlp.down_proj": {"inputs": [[[0.87109375]]], "outputs": [[[0.328125]], [[0.89453125]]], "params": {"weight": [[0.2890625]]}}, "model.layers.14.self_attn.qkv_proj": {"inputs": [[[6.4375]]], "params": {"weight": [[0.287109375]]}}, "model.layers.14.self_attn.o_proj": {"inputs": [[[0.921875]]], "outputs": [[[0.27734375]], [[0.490234375]]], "params": {"weight": [[0.15625]]}}, "model.layers.14.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.68359375]], [[15.9375]]]}, "model.layers.14.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.1953125]]]}, "model.layers.14.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[19.5]]]}, "model.layers.14.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.0]]]}, "model.layers.14.self_attn.attn.impl.k_cache": {"inputs": [[[15.9375]]]}, "model.layers.14.self_attn.attn.impl.v_cache": {"inputs": [[[1.1953125]]]}, "model.layers.14.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.90625]], [[15.4375]], [[1.1015625]]], "outputs": [[[0.69140625]], [[1.0]]]}, "model.layers.14.mlp.gate_up_proj": {"inputs": [[[3.875]]], "params": {"weight": [[0.21875]]}}, "model.layers.14.mlp.down_proj": {"inputs": [[[0.890625]]], "outputs": [[[0.306640625]], [[0.87109375]]], "params": {"weight": [[0.2353515625]]}}, "model.layers.15.self_attn.qkv_proj": {"inputs": [[[7.0]]], "params": {"weight": [[0.30859375]]}}, "model.layers.15.self_attn.o_proj": {"inputs": [[[0.57421875]]], "outputs": [[[0.1884765625]], [[0.4140625]]], "params": {"weight": [[0.20703125]]}}, "model.layers.15.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.55078125]], [[10.0]]]}, "model.layers.15.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.1328125]]]}, "model.layers.15.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[32.75]]]}, "model.layers.15.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[19.875]]]}, "model.layers.15.self_attn.attn.impl.k_cache": {"inputs": [[[10.0]]]}, "model.layers.15.self_attn.attn.impl.v_cache": {"inputs": [[[1.1328125]]]}, "model.layers.15.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.96875]], [[9.4375]], [[0.9921875]]], "outputs": [[[0.310546875]], [[1.0]]]}, "model.layers.15.mlp.gate_up_proj": {"inputs": [[[3.453125]]], "params": {"weight": [[0.7265625]]}}, "model.layers.15.mlp.down_proj": {"inputs": [[[0.91015625]]], "outputs": [[[0.11328125]], [[0.375]]], "params": {"weight": [[0.455078125]]}}, "model.layers.16.self_attn.qkv_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.263671875]]}}, "model.layers.16.self_attn.o_proj": {"inputs": [[[0.4453125]]], "outputs": [[[0.263671875]], [[0.5234375]]], "params": {"weight": [[0.189453125]]}}, "model.layers.16.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6640625]], [[15.6875]]]}, "model.layers.16.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.25]]]}, "model.layers.16.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[23.25]]]}, "model.layers.16.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[11.25]]]}, "model.layers.16.self_attn.attn.impl.k_cache": {"inputs": [[[15.6875]]]}, "model.layers.16.self_attn.attn.impl.v_cache": {"inputs": [[[1.25]]]}, "model.layers.16.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.1875]], [[14.8125]], [[1.015625]]], "outputs": [[[0.2470703125]], [[1.0]]]}, "model.layers.16.mlp.gate_up_proj": {"inputs": [[[4.1875]]], "params": {"weight": [[1.0703125]]}}, "model.layers.16.mlp.down_proj": {"inputs": [[[2.03125]]], "outputs": [[[0.32421875]], [[0.6328125]]], "params": {"weight": [[0.490234375]]}}, "model.layers.17.self_attn.qkv_proj": {"inputs": [[[10.9375]]], "params": {"weight": [[0.306640625]]}}, "model.layers.17.self_attn.o_proj": {"inputs": [[[0.53515625]]], "outputs": [[[0.236328125]], [[0.75]]], "params": {"weight": [[0.1884765625]]}}, "model.layers.17.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.62890625]], [[12.75]]]}, "model.layers.17.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.0234375]]]}, "model.layers.17.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[18.875]]]}, "model.layers.17.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[12.375]]]}, "model.layers.17.self_attn.attn.impl.k_cache": {"inputs": [[[12.75]]]}, "model.layers.17.self_attn.attn.impl.v_cache": {"inputs": [[[1.0234375]]]}, "model.layers.17.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.625]], [[11.125]], [[1.0234375]]], "outputs": [[[0.515625]], [[1.0]]]}, "model.layers.17.mlp.gate_up_proj": {"inputs": [[[4.03125]]], "params": {"weight": [[0.283203125]]}}, "model.layers.17.mlp.down_proj": {"inputs": [[[1.5234375]]], "outputs": [[[0.2353515625]], [[0.84765625]]], "params": {"weight": [[0.78125]]}}, "model.layers.18.self_attn.qkv_proj": {"inputs": [[[11.25]]], "params": {"weight": [[0.2431640625]]}}, "model.layers.18.self_attn.o_proj": {"inputs": [[[0.55859375]]], "outputs": [[[0.20703125]], [[0.640625]]], "params": {"weight": [[0.2353515625]]}}, "model.layers.18.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.65625]], [[12.25]]]}, "model.layers.18.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.2890625]]]}, "model.layers.18.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[30.5]]]}, "model.layers.18.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[17.125]]]}, "model.layers.18.self_attn.attn.impl.k_cache": {"inputs": [[[12.25]]]}, "model.layers.18.self_attn.attn.impl.v_cache": {"inputs": [[[1.2890625]]]}, "model.layers.18.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.59375]], [[11.125]], [[1.1640625]]], "outputs": [[[0.296875]], [[1.0]]]}, "model.layers.18.mlp.gate_up_proj": {"inputs": [[[7.5625]]], "params": {"weight": [[0.3203125]]}}, "model.layers.18.mlp.down_proj": {"inputs": [[[1.96875]]], "outputs": [[[6.625]], [[30.625]]], "params": {"weight": [[0.6015625]]}}, "model.layers.19.self_attn.qkv_proj": {"inputs": [[[10.6875]]], "params": {"weight": [[0.2490234375]]}}, "model.layers.19.self_attn.o_proj": {"inputs": [[[0.435546875]]], "outputs": [[[0.28515625]], [[0.76171875]]], "params": {"weight": [[0.373046875]]}}, "model.layers.19.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.67578125]], [[13.5]]]}, "model.layers.19.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.140625]]]}, "model.layers.19.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[39.0]]]}, "model.layers.19.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.875]]]}, "model.layers.19.self_attn.attn.impl.k_cache": {"inputs": [[[13.5]]]}, "model.layers.19.self_attn.attn.impl.v_cache": {"inputs": [[[1.140625]]]}, "model.layers.19.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.375]], [[12.375]], [[1.140625]]], "outputs": [[[0.373046875]], [[1.0]]]}, "model.layers.19.mlp.gate_up_proj": {"inputs": [[[4.09375]]], "params": {"weight": [[0.2490234375]]}}, "model.layers.19.mlp.down_proj": {"inputs": [[[2.234375]]], "outputs": [[[0.44921875]], [[1.2265625]]], "params": {"weight": [[0.39453125]]}}, "model.layers.20.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.251953125]]}}, "model.layers.20.self_attn.o_proj": {"inputs": [[[0.400390625]]], "outputs": [[[0.2451171875]], [[1.1484375]]], "params": {"weight": [[0.42578125]]}}, "model.layers.20.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.55859375]], [[11.25]]]}, "model.layers.20.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.6875]]]}, "model.layers.20.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.375]]]}, "model.layers.20.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.40625]]]}, "model.layers.20.self_attn.attn.impl.k_cache": {"inputs": [[[11.25]]]}, "model.layers.20.self_attn.attn.impl.v_cache": {"inputs": [[[1.6875]]]}, "model.layers.20.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.8125]], [[10.5625]], [[1.640625]]], "outputs": [[[0.400390625]], [[1.0]]]}, "model.layers.20.mlp.gate_up_proj": {"inputs": [[[3.578125]]], "params": {"weight": [[0.3359375]]}}, "model.layers.20.mlp.down_proj": {"inputs": [[[2.625]]], "outputs": [[[0.73046875]], [[4.6875]]], "params": {"weight": [[0.478515625]]}}, "model.layers.21.self_attn.qkv_proj": {"inputs": [[[11.4375]]], "params": {"weight": [[0.30859375]]}}, "model.layers.21.self_attn.o_proj": {"inputs": [[[0.68359375]]], "outputs": [[[0.2890625]], [[1.3984375]]], "params": {"weight": [[0.1357421875]]}}, "model.layers.21.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.77734375]], [[15.4375]]]}, "model.layers.21.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.65625]]]}, "model.layers.21.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[32.25]]]}, "model.layers.21.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[19.25]]]}, "model.layers.21.self_attn.attn.impl.k_cache": {"inputs": [[[15.4375]]]}, "model.layers.21.self_attn.attn.impl.v_cache": {"inputs": [[[2.65625]]]}, "model.layers.21.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.84375]], [[15.1875]], [[2.65625]]], "outputs": [[[0.66796875]], [[1.0]]]}, "model.layers.21.mlp.gate_up_proj": {"inputs": [[[3.578125]]], "params": {"weight": [[0.37109375]]}}, "model.layers.21.mlp.down_proj": {"inputs": [[[1.90625]]], "outputs": [[[0.68359375]], [[5.0]]], "params": {"weight": [[0.88671875]]}}, "model.layers.22.self_attn.qkv_proj": {"inputs": [[[11.0]]], "params": {"weight": [[0.3125]]}}, "model.layers.22.self_attn.o_proj": {"inputs": [[[0.41796875]]], "outputs": [[[0.37109375]], [[1.25]]], "params": {"weight": [[0.388671875]]}}, "model.layers.22.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6328125]], [[15.25]]]}, "model.layers.22.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.03125]]]}, "model.layers.22.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[15.125]]]}, "model.layers.22.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.625]]]}, "model.layers.22.self_attn.attn.impl.k_cache": {"inputs": [[[15.25]]]}, "model.layers.22.self_attn.attn.impl.v_cache": {"inputs": [[[1.03125]]]}, "model.layers.22.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.46875]], [[13.6875]], [[0.9140625]]], "outputs": [[[0.41796875]], [[1.0]]]}, "model.layers.22.mlp.gate_up_proj": {"inputs": [[[3.25]]], "params": {"weight": [[0.400390625]]}}, "model.layers.22.mlp.down_proj": {"inputs": [[[4.46875]]], "outputs": [[[0.478515625]], [[2.578125]]], "params": {"weight": [[0.578125]]}}, "model.layers.23.self_attn.qkv_proj": {"inputs": [[[11.6875]]], "params": {"weight": [[0.27734375]]}}, "model.layers.23.self_attn.o_proj": {"inputs": [[[1.0703125]]], "outputs": [[[0.578125]], [[2.140625]]], "params": {"weight": [[0.447265625]]}}, "model.layers.23.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.67578125]], [[13.0]]]}, "model.layers.23.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[6.125]]]}, "model.layers.23.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[14.0]]]}, "model.layers.23.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.125]]]}, "model.layers.23.self_attn.attn.impl.k_cache": {"inputs": [[[13.0]]]}, "model.layers.23.self_attn.attn.impl.v_cache": {"inputs": [[[6.125]]]}, "model.layers.23.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.84375]], [[13.0]], [[6.125]]], "outputs": [[[1.0703125]], [[1.0]]]}, "model.layers.23.mlp.gate_up_proj": {"inputs": [[[3.265625]]], "params": {"weight": [[0.59375]]}}, "model.layers.23.mlp.down_proj": {"inputs": [[[2.5625]]], "outputs": [[[0.5625]], [[2.75]]], "params": {"weight": [[0.275390625]]}}, "model.layers.24.self_attn.qkv_proj": {"inputs": [[[9.3125]]], "params": {"weight": [[0.45703125]]}}, "model.layers.24.self_attn.o_proj": {"inputs": [[[0.37890625]]], "outputs": [[[0.2158203125]], [[2.0]]], "params": {"weight": [[0.267578125]]}}, "model.layers.24.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.69140625]], [[14.3125]]]}, "model.layers.24.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.625]]]}, "model.layers.24.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.4375]]]}, "model.layers.24.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.9375]]]}, "model.layers.24.self_attn.attn.impl.k_cache": {"inputs": [[[14.3125]]]}, "model.layers.24.self_attn.attn.impl.v_cache": {"inputs": [[[3.625]]]}, "model.layers.24.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.25]], [[12.0]], [[3.625]]], "outputs": [[[0.240234375]], [[1.0]]]}, "model.layers.24.mlp.gate_up_proj": {"inputs": [[[3.4375]]], "params": {"weight": [[0.546875]]}}, "model.layers.24.mlp.down_proj": {"inputs": [[[1.8125]]], "outputs": [[[2.078125]], [[14.9375]]], "params": {"weight": [[0.640625]]}}, "model.layers.25.self_attn.qkv_proj": {"inputs": [[[10.5625]]], "params": {"weight": [[0.26953125]]}}, "model.layers.25.self_attn.o_proj": {"inputs": [[[0.73828125]]], "outputs": [[[0.375]], [[2.28125]]], "params": {"weight": [[0.451171875]]}}, "model.layers.25.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.76953125]], [[12.625]]]}, "model.layers.25.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.3125]]]}, "model.layers.25.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.59375]]]}, "model.layers.25.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.4375]]]}, "model.layers.25.self_attn.attn.impl.k_cache": {"inputs": [[[12.625]]]}, "model.layers.25.self_attn.attn.impl.v_cache": {"inputs": [[[2.3125]]]}, "model.layers.25.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.0]], [[12.3125]], [[2.3125]]], "outputs": [[[0.62109375]], [[1.0]]]}, "model.layers.25.mlp.gate_up_proj": {"inputs": [[[3.5]]], "params": {"weight": [[0.40625]]}}, "model.layers.25.mlp.down_proj": {"inputs": [[[2.609375]]], "outputs": [[[0.9375]], [[6.28125]]], "params": {"weight": [[0.53125]]}}, "model.layers.26.self_attn.qkv_proj": {"inputs": [[[10.875]]], "params": {"weight": [[0.388671875]]}}, "model.layers.26.self_attn.o_proj": {"inputs": [[[1.2109375]]], "outputs": [[[0.341796875]], [[2.1875]]], "params": {"weight": [[0.546875]]}}, "model.layers.26.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.9296875]], [[17.375]]]}, "model.layers.26.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[5.15625]]]}, "model.layers.26.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[38.5]]]}, "model.layers.26.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.1875]]]}, "model.layers.26.self_attn.attn.impl.k_cache": {"inputs": [[[17.375]]]}, "model.layers.26.self_attn.attn.impl.v_cache": {"inputs": [[[5.15625]]]}, "model.layers.26.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.90625]], [[15.5]], [[5.15625]]], "outputs": [[[0.91015625]], [[1.0]]]}, "model.layers.26.mlp.gate_up_proj": {"inputs": [[[3.890625]]], "params": {"weight": [[0.287109375]]}}, "model.layers.26.mlp.down_proj": {"inputs": [[[1.6171875]]], "outputs": [[[0.984375]], [[3.546875]]], "params": {"weight": [[0.7109375]]}}, "model.layers.27.self_attn.qkv_proj": {"inputs": [[[11.0]]], "params": {"weight": [[0.328125]]}}, "model.layers.27.self_attn.o_proj": {"inputs": [[[0.734375]]], "outputs": [[[0.9921875]], [[2.34375]]], "params": {"weight": [[0.4140625]]}}, "model.layers.27.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.73046875]], [[18.375]]]}, "model.layers.27.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.28125]]]}, "model.layers.27.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[47.0]]]}, "model.layers.27.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[17.0]]]}, "model.layers.27.self_attn.attn.impl.k_cache": {"inputs": [[[18.375]]]}, "model.layers.27.self_attn.attn.impl.v_cache": {"inputs": [[[1.28125]]]}, "model.layers.27.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.78125]], [[18.125]], [[1.28125]]], "outputs": [[[0.47265625]], [[1.0]]]}, "model.layers.27.mlp.gate_up_proj": {"inputs": [[[4.0625]]], "params": {"weight": [[0.447265625]]}}, "model.layers.27.mlp.down_proj": {"inputs": [[[2.3125]]], "outputs": [[[0.85546875]], [[2.375]]], "params": {"weight": [[0.9609375]]}}, "model.layers.28.self_attn.qkv_proj": {"inputs": [[[11.0]]], "params": {"weight": [[0.267578125]]}}, "model.layers.28.self_attn.o_proj": {"inputs": [[[0.55078125]]], "outputs": [[[0.82421875]], [[2.625]]], "params": {"weight": [[0.396484375]]}}, "model.layers.28.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.76171875]], [[19.0]]]}, "model.layers.28.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.28125]]]}, "model.layers.28.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[29.625]]]}, "model.layers.28.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[17.875]]]}, "model.layers.28.self_attn.attn.impl.k_cache": {"inputs": [[[19.0]]]}, "model.layers.28.self_attn.attn.impl.v_cache": {"inputs": [[[3.28125]]]}, "model.layers.28.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.0625]], [[17.375]], [[3.28125]]], "outputs": [[[0.55078125]], [[1.0]]]}, "model.layers.28.mlp.gate_up_proj": {"inputs": [[[4.78125]]], "params": {"weight": [[0.341796875]]}}, "model.layers.28.mlp.down_proj": {"inputs": [[[1.765625]]], "outputs": [[[0.8984375]], [[2.90625]]], "params": {"weight": [[0.62109375]]}}, "model.layers.29.self_attn.qkv_proj": {"inputs": [[[11.125]]], "params": {"weight": [[0.3515625]]}}, "model.layers.29.self_attn.o_proj": {"inputs": [[[0.546875]]], "outputs": [[[2.171875]], [[2.625]]], "params": {"weight": [[0.23046875]]}}, "model.layers.29.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.8125]], [[19.625]]]}, "model.layers.29.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.046875]]]}, "model.layers.29.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[45.0]]]}, "model.layers.29.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[18.75]]]}, "model.layers.29.self_attn.attn.impl.k_cache": {"inputs": [[[19.625]]]}, "model.layers.29.self_attn.attn.impl.v_cache": {"inputs": [[[3.046875]]]}, "model.layers.29.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.53125]], [[16.25]], [[3.046875]]], "outputs": [[[0.5234375]], [[1.0]]]}, "model.layers.29.mlp.gate_up_proj": {"inputs": [[[4.9375]]], "params": {"weight": [[0.326171875]]}}, "model.layers.29.mlp.down_proj": {"inputs": [[[1.515625]]], "outputs": [[[1.046875]], [[2.390625]]], "params": {"weight": [[0.71875]]}}, "model.layers.30.self_attn.qkv_proj": {"inputs": [[[10.25]]], "params": {"weight": [[0.35546875]]}}, "model.layers.30.self_attn.o_proj": {"inputs": [[[0.6015625]]], "outputs": [[[0.38671875]], [[2.03125]]], "params": {"weight": [[0.57421875]]}}, "model.layers.30.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.78515625]], [[16.5]]]}, "model.layers.30.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[8.0625]]]}, "model.layers.30.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[31.375]]]}, "model.layers.30.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[17.0]]]}, "model.layers.30.self_attn.attn.impl.k_cache": {"inputs": [[[16.5]]]}, "model.layers.30.self_attn.attn.impl.v_cache": {"inputs": [[[8.0625]]]}, "model.layers.30.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.375]], [[15.375]], [[8.0625]]], "outputs": [[[0.6015625]], [[1.0]]]}, "model.layers.30.mlp.gate_up_proj": {"inputs": [[[6.5]]], "params": {"weight": [[0.34375]]}}, "model.layers.30.mlp.down_proj": {"inputs": [[[2.640625]]], "outputs": [[[0.7890625]], [[3.375]]], "params": {"weight": [[0.74609375]]}}, "model.layers.31.self_attn.qkv_proj": {"inputs": [[[12.5]]], "params": {"weight": [[0.2373046875]]}}, "model.layers.31.self_attn.o_proj": {"inputs": [[[0.96484375]]], "outputs": [[[1.4921875]], [[2.265625]]], "params": {"weight": [[0.224609375]]}}, "model.layers.31.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.87890625]], [[13.9375]]]}, "model.layers.31.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.5]]]}, "model.layers.31.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[32.75]]]}, "model.layers.31.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.625]]]}, "model.layers.31.self_attn.attn.impl.k_cache": {"inputs": [[[13.9375]]]}, "model.layers.31.self_attn.attn.impl.v_cache": {"inputs": [[[1.5]]]}, "model.layers.31.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.3125]], [[13.25]], [[1.3984375]]], "outputs": [[[0.53125]], [[1.0]]]}, "model.layers.31.mlp.gate_up_proj": {"inputs": [[[5.09375]]], "params": {"weight": [[0.455078125]]}}, "model.layers.31.mlp.down_proj": {"inputs": [[[2.4375]]], "outputs": [[[0.60546875]], [[1.5390625]]], "params": {"weight": [[0.69140625]]}}, "model.layers.32.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.296875]]}}, "model.layers.32.self_attn.o_proj": {"inputs": [[[0.546875]]], "outputs": [[[0.470703125]], [[1.5]]], "params": {"weight": [[0.5859375]]}}, "model.layers.32.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.80078125]], [[17.375]]]}, "model.layers.32.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.6796875]]]}, "model.layers.32.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[40.5]]]}, "model.layers.32.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[19.75]]]}, "model.layers.32.self_attn.attn.impl.k_cache": {"inputs": [[[17.375]]]}, "model.layers.32.self_attn.attn.impl.v_cache": {"inputs": [[[1.6796875]]]}, "model.layers.32.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.53125]], [[16.75]], [[1.6796875]]], "outputs": [[[0.373046875]], [[1.0]]]}, "model.layers.32.mlp.gate_up_proj": {"inputs": [[[6.65625]]], "params": {"weight": [[0.36328125]]}}, "model.layers.32.mlp.down_proj": {"inputs": [[[4.375]]], "outputs": [[[0.703125]], [[2.703125]]], "params": {"weight": [[0.640625]]}}, "model.layers.33.self_attn.qkv_proj": {"inputs": [[[14.25]]], "params": {"weight": [[0.357421875]]}}, "model.layers.33.self_attn.o_proj": {"inputs": [[[1.03125]]], "outputs": [[[1.4921875]], [[2.65625]]], "params": {"weight": [[0.453125]]}}, "model.layers.33.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.87109375]], [[18.25]]]}, "model.layers.33.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.8359375]]]}, "model.layers.33.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[29.25]]]}, "model.layers.33.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.9375]]]}, "model.layers.33.self_attn.attn.impl.k_cache": {"inputs": [[[18.25]]]}, "model.layers.33.self_attn.attn.impl.v_cache": {"inputs": [[[1.8359375]]]}, "model.layers.33.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.625]], [[15.1875]], [[1.8359375]]], "outputs": [[[0.734375]], [[1.0]]]}, "model.layers.33.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[3.421875]]}}, "model.layers.33.mlp.down_proj": {"inputs": [[[1.8828125]]], "outputs": [[[0.890625]], [[1.7578125]]], "params": {"weight": [[0.6796875]]}}, "model.layers.34.self_attn.qkv_proj": {"inputs": [[[13.25]]], "params": {"weight": [[0.365234375]]}}, "model.layers.34.self_attn.o_proj": {"inputs": [[[0.59765625]]], "outputs": [[[13.5625]], [[14.9375]]], "params": {"weight": [[0.271484375]]}}, "model.layers.34.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.8515625]], [[16.0]]]}, "model.layers.34.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.71875]]]}, "model.layers.34.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[24.125]]]}, "model.layers.34.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.75]]]}, "model.layers.34.self_attn.attn.impl.k_cache": {"inputs": [[[16.0]]]}, "model.layers.34.self_attn.attn.impl.v_cache": {"inputs": [[[2.71875]]]}, "model.layers.34.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.5625]], [[14.75]], [[2.71875]]], "outputs": [[[0.54296875]], [[1.0]]]}, "model.layers.34.mlp.gate_up_proj": {"inputs": [[[7.0]]], "params": {"weight": [[0.373046875]]}}, "model.layers.34.mlp.down_proj": {"inputs": [[[2.03125]]], "outputs": [[[0.85546875]], [[3.125]]], "params": {"weight": [[1.3359375]]}}, "model.layers.35.self_attn.qkv_proj": {"inputs": [[[11.8125]]], "params": {"weight": [[0.3203125]]}}, "model.layers.35.self_attn.o_proj": {"inputs": [[[0.69140625]]], "outputs": [[[5.90625]], [[6.71875]]], "params": {"weight": [[0.19921875]]}}, "model.layers.35.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.84375]], [[16.375]]]}, "model.layers.35.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.125]]]}, "model.layers.35.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[52.25]]]}, "model.layers.35.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[18.25]]]}, "model.layers.35.self_attn.attn.impl.k_cache": {"inputs": [[[16.375]]]}, "model.layers.35.self_attn.attn.impl.v_cache": {"inputs": [[[1.125]]]}, "model.layers.35.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.875]], [[15.375]], [[1.1015625]]], "outputs": [[[0.5546875]], [[1.0]]]}, "model.layers.35.mlp.gate_up_proj": {"inputs": [[[6.5]]], "params": {"weight": [[0.427734375]]}}, "model.layers.35.mlp.down_proj": {"inputs": [[[5.1875]]], "outputs": [[[0.8125]], [[4.46875]]], "params": {"weight": [[1.2734375]]}}, "model.layers.36.self_attn.qkv_proj": {"inputs": [[[15.4375]]], "params": {"weight": [[0.330078125]]}}, "model.layers.36.self_attn.o_proj": {"inputs": [[[0.578125]]], "outputs": [[[2.71875]], [[2.75]]], "params": {"weight": [[0.265625]]}}, "model.layers.36.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.62890625]], [[14.0]]]}, "model.layers.36.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.6328125]]]}, "model.layers.36.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[35.5]]]}, "model.layers.36.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.25]]]}, "model.layers.36.self_attn.attn.impl.k_cache": {"inputs": [[[14.0]]]}, "model.layers.36.self_attn.attn.impl.v_cache": {"inputs": [[[1.6328125]]]}, "model.layers.36.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.21875]], [[13.6875]], [[1.6328125]]], "outputs": [[[0.578125]], [[1.0]]]}, "model.layers.36.mlp.gate_up_proj": {"inputs": [[[8.125]]], "params": {"weight": [[0.703125]]}}, "model.layers.36.mlp.down_proj": {"inputs": [[[2.5]]], "outputs": [[[0.73828125]], [[3.40625]]], "params": {"weight": [[0.67578125]]}}, "model.layers.37.self_attn.qkv_proj": {"inputs": [[[17.625]]], "params": {"weight": [[0.373046875]]}}, "model.layers.37.self_attn.o_proj": {"inputs": [[[0.62109375]]], "outputs": [[[2.859375]], [[3.59375]]], "params": {"weight": [[0.1337890625]]}}, "model.layers.37.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.7578125]], [[16.5]]]}, "model.layers.37.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.015625]]]}, "model.layers.37.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[45.0]]]}, "model.layers.37.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[18.25]]]}, "model.layers.37.self_attn.attn.impl.k_cache": {"inputs": [[[16.5]]]}, "model.layers.37.self_attn.attn.impl.v_cache": {"inputs": [[[1.015625]]]}, "model.layers.37.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.375]], [[16.125]], [[1.015625]]], "outputs": [[[0.56640625]], [[1.0]]]}, "model.layers.37.mlp.gate_up_proj": {"inputs": [[[7.4375]]], "params": {"weight": [[0.359375]]}}, "model.layers.37.mlp.down_proj": {"inputs": [[[3.421875]]], "outputs": [[[0.78515625]], [[2.21875]]], "params": {"weight": [[0.400390625]]}}, "model.layers.38.self_attn.qkv_proj": {"inputs": [[[16.125]]], "params": {"weight": [[0.38671875]]}}, "model.layers.38.self_attn.o_proj": {"inputs": [[[0.78125]]], "outputs": [[[0.703125]], [[1.109375]]], "params": {"weight": [[0.166015625]]}}, "model.layers.38.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.828125]], [[16.125]]]}, "model.layers.38.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.1796875]]]}, "model.layers.38.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[35.75]]]}, "model.layers.38.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.0]]]}, "model.layers.38.self_attn.attn.impl.k_cache": {"inputs": [[[16.125]]]}, "model.layers.38.self_attn.attn.impl.v_cache": {"inputs": [[[1.1796875]]]}, "model.layers.38.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.3125]], [[14.6875]], [[1.1796875]]], "outputs": [[[0.53515625]], [[1.0]]]}, "model.layers.38.mlp.gate_up_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.53125]]}}, "model.layers.38.mlp.down_proj": {"inputs": [[[2.53125]]], "outputs": [[[2.140625]], [[3.078125]]], "params": {"weight": [[0.921875]]}}, "model.layers.39.self_attn.qkv_proj": {"inputs": [[[17.25]]], "params": {"weight": [[0.58984375]]}}, "model.layers.39.self_attn.o_proj": {"inputs": [[[0.55859375]]], "outputs": [[[3.515625]], [[3.625]]], "params": {"weight": [[0.9609375]]}}, "model.layers.39.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6171875]], [[13.4375]]]}, "model.layers.39.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.59375]]]}, "model.layers.39.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[6.125]]]}, "model.layers.39.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.375]]]}, "model.layers.39.self_attn.attn.impl.k_cache": {"inputs": [[[13.4375]]]}, "model.layers.39.self_attn.attn.impl.v_cache": {"inputs": [[[4.59375]]]}, "model.layers.39.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.59375]], [[12.6875]], [[4.59375]]], "outputs": [[[0.416015625]], [[1.0]]]}, "model.layers.39.mlp.gate_up_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.412109375]]}}, "model.layers.39.mlp.down_proj": {"inputs": [[[4.03125]]], "outputs": [[[1.6328125]], [[3.203125]]], "params": {"weight": [[1.6484375]]}}, "model.layers.40.self_attn.qkv_proj": {"inputs": [[[13.0]]], "params": {"weight": [[0.33203125]]}}, "model.layers.40.self_attn.o_proj": {"inputs": [[[0.82421875]]], "outputs": [[[0.9609375]], [[2.0625]]], "params": {"weight": [[0.56640625]]}}, "model.layers.40.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.60546875]], [[11.125]]]}, "model.layers.40.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.4375]]]}, "model.layers.40.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[6.90625]]]}, "model.layers.40.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.84375]]]}, "model.layers.40.self_attn.attn.impl.k_cache": {"inputs": [[[11.125]]]}, "model.layers.40.self_attn.attn.impl.v_cache": {"inputs": [[[4.4375]]]}, "model.layers.40.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.59375]], [[10.4375]], [[4.4375]]], "outputs": [[[0.51171875]], [[1.0]]]}, "model.layers.40.mlp.gate_up_proj": {"inputs": [[[7.1875]]], "params": {"weight": [[0.62890625]]}}, "model.layers.40.mlp.down_proj": {"inputs": [[[1.921875]]], "outputs": [[[0.68359375]], [[1.34375]]], "params": {"weight": [[0.283203125]]}}, "model.layers.41.self_attn.qkv_proj": {"inputs": [[[16.25]]], "params": {"weight": [[0.33203125]]}}, "model.layers.41.self_attn.o_proj": {"inputs": [[[1.1484375]]], "outputs": [[[0.2255859375]], [[1.078125]]], "params": {"weight": [[0.328125]]}}, "model.layers.41.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.8046875]], [[16.25]]]}, "model.layers.41.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.5625]]]}, "model.layers.41.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[34.25]]]}, "model.layers.41.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[12.0]]]}, "model.layers.41.self_attn.attn.impl.k_cache": {"inputs": [[[16.25]]]}, "model.layers.41.self_attn.attn.impl.v_cache": {"inputs": [[[1.5625]]]}, "model.layers.41.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.0]], [[15.0625]], [[1.0390625]]], "outputs": [[[0.765625]], [[1.0]]]}, "model.layers.41.mlp.gate_up_proj": {"inputs": [[[7.28125]]], "params": {"weight": [[0.32421875]]}}, "model.layers.41.mlp.down_proj": {"inputs": [[[4.15625]]], "outputs": [[[0.9921875]], [[1.3984375]]], "params": {"weight": [[1.375]]}}, "model.layers.42.self_attn.qkv_proj": {"inputs": [[[14.8125]]], "params": {"weight": [[0.455078125]]}}, "model.layers.42.self_attn.o_proj": {"inputs": [[[0.6953125]]], "outputs": [[[0.1474609375]], [[0.5]]], "params": {"weight": [[0.3359375]]}}, "model.layers.42.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.7109375]], [[16.875]]]}, "model.layers.42.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.390625]]]}, "model.layers.42.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[39.0]]]}, "model.layers.42.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.1875]]]}, "model.layers.42.self_attn.attn.impl.k_cache": {"inputs": [[[16.875]]]}, "model.layers.42.self_attn.attn.impl.v_cache": {"inputs": [[[1.390625]]]}, "model.layers.42.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.0]], [[15.1875]], [[1.2578125]]], "outputs": [[[0.6953125]], [[1.0]]]}, "model.layers.42.mlp.gate_up_proj": {"inputs": [[[6.75]]], "params": {"weight": [[0.392578125]]}}, "model.layers.42.mlp.down_proj": {"inputs": [[[3.078125]]], "outputs": [[[0.85546875]], [[2.046875]]], "params": {"weight": [[0.271484375]]}}, "model.layers.43.self_attn.qkv_proj": {"inputs": [[[12.6875]]], "params": {"weight": [[0.44921875]]}}, "model.layers.43.self_attn.o_proj": {"inputs": [[[0.78125]]], "outputs": [[[0.357421875]], [[0.734375]]], "params": {"weight": [[0.61328125]]}}, "model.layers.43.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.65625]], [[17.25]]]}, "model.layers.43.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[6.375]]]}, "model.layers.43.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[21.375]]]}, "model.layers.43.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.875]]]}, "model.layers.43.self_attn.attn.impl.k_cache": {"inputs": [[[17.25]]]}, "model.layers.43.self_attn.attn.impl.v_cache": {"inputs": [[[6.375]]]}, "model.layers.43.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.78125]], [[15.125]], [[6.375]]], "outputs": [[[0.76953125]], [[1.0]]]}, "model.layers.43.mlp.gate_up_proj": {"inputs": [[[6.625]]], "params": {"weight": [[0.353515625]]}}, "model.layers.43.mlp.down_proj": {"inputs": [[[3.921875]]], "outputs": [[[0.60546875]], [[0.81640625]]], "params": {"weight": [[0.494140625]]}}, "model.layers.44.self_attn.qkv_proj": {"inputs": [[[13.625]]], "params": {"weight": [[0.298828125]]}}, "model.layers.44.self_attn.o_proj": {"inputs": [[[0.6796875]]], "outputs": [[[0.67578125]], [[0.59375]]], "params": {"weight": [[0.162109375]]}}, "model.layers.44.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.640625]], [[14.0625]]]}, "model.layers.44.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.640625]]]}, "model.layers.44.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.28125]]]}, "model.layers.44.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.75]]]}, "model.layers.44.self_attn.attn.impl.k_cache": {"inputs": [[[14.0625]]]}, "model.layers.44.self_attn.attn.impl.v_cache": {"inputs": [[[1.640625]]]}, "model.layers.44.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.59375]], [[13.75]], [[1.640625]]], "outputs": [[[0.5390625]], [[1.0]]]}, "model.layers.44.mlp.gate_up_proj": {"inputs": [[[7.09375]]], "params": {"weight": [[0.5078125]]}}, "model.layers.44.mlp.down_proj": {"inputs": [[[3.875]]], "outputs": [[[0.486328125]], [[0.984375]]], "params": {"weight": [[0.90234375]]}}, "model.layers.45.self_attn.qkv_proj": {"inputs": [[[13.75]]], "params": {"weight": [[0.462890625]]}}, "model.layers.45.self_attn.o_proj": {"inputs": [[[0.6171875]]], "outputs": [[[0.546875]], [[0.9921875]]], "params": {"weight": [[0.375]]}}, "model.layers.45.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.7109375]], [[12.75]]]}, "model.layers.45.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.71875]]]}, "model.layers.45.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.0]]]}, "model.layers.45.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.4375]]]}, "model.layers.45.self_attn.attn.impl.k_cache": {"inputs": [[[12.75]]]}, "model.layers.45.self_attn.attn.impl.v_cache": {"inputs": [[[4.71875]]]}, "model.layers.45.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.5625]], [[11.5625]], [[4.71875]]], "outputs": [[[0.357421875]], [[1.0]]]}, "model.layers.45.mlp.gate_up_proj": {"inputs": [[[7.15625]]], "params": {"weight": [[0.58984375]]}}, "model.layers.45.mlp.down_proj": {"inputs": [[[3.78125]]], "outputs": [[[1.515625]], [[2.4375]]], "params": {"weight": [[0.326171875]]}}, "model.layers.46.self_attn.qkv_proj": {"inputs": [[[12.9375]]], "params": {"weight": [[0.69921875]]}}, "model.layers.46.self_attn.o_proj": {"inputs": [[[0.451171875]]], "outputs": [[[0.14453125]], [[0.2099609375]]], "params": {"weight": [[0.306640625]]}}, "model.layers.46.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.84375]], [[15.5625]]]}, "model.layers.46.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.21875]]]}, "model.layers.46.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[9.1875]]]}, "model.layers.46.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.84375]]]}, "model.layers.46.self_attn.attn.impl.k_cache": {"inputs": [[[15.5625]]]}, "model.layers.46.self_attn.attn.impl.v_cache": {"inputs": [[[3.21875]]]}, "model.layers.46.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.9375]], [[15.5625]], [[3.21875]]], "outputs": [[[0.349609375]], [[1.0]]]}, "model.layers.46.mlp.gate_up_proj": {"inputs": [[[6.4375]]], "params": {"weight": [[0.52734375]]}}, "model.layers.46.mlp.down_proj": {"inputs": [[[3.125]]], "outputs": [[[0.359375]], [[0.671875]]], "params": {"weight": [[0.21875]]}}, "model.layers.47.self_attn.qkv_proj": {"inputs": [[[11.9375]]], "params": {"weight": [[0.4296875]]}}, "model.layers.47.self_attn.o_proj": {"inputs": [[[0.9296875]]], "outputs": [[[0.625]], [[0.64453125]]], "params": {"weight": [[0.427734375]]}}, "model.layers.47.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.89453125]], [[14.0625]]]}, "model.layers.47.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.359375]]]}, "model.layers.47.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[21.75]]]}, "model.layers.47.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.3125]]]}, "model.layers.47.self_attn.attn.impl.k_cache": {"inputs": [[[14.0625]]]}, "model.layers.47.self_attn.attn.impl.v_cache": {"inputs": [[[1.359375]]]}, "model.layers.47.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.5625]], [[14.0625]], [[1.359375]]], "outputs": [[[0.7265625]], [[1.0]]]}, "model.layers.47.mlp.gate_up_proj": {"inputs": [[[6.5625]]], "params": {"weight": [[0.478515625]]}}, "model.layers.47.mlp.down_proj": {"inputs": [[[2.8125]]], "outputs": [[[0.5234375]], [[1.21875]]], "params": {"weight": [[0.275390625]]}}, "model.layers.48.self_attn.qkv_proj": {"inputs": [[[11.125]]], "params": {"weight": [[0.421875]]}}, "model.layers.48.self_attn.o_proj": {"inputs": [[[0.65625]]], "outputs": [[[0.3984375]], [[0.515625]]], "params": {"weight": [[0.275390625]]}}, "model.layers.48.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.42578125]], [[12.8125]]]}, "model.layers.48.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.359375]]]}, "model.layers.48.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[10.375]]]}, "model.layers.48.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.59375]]]}, "model.layers.48.self_attn.attn.impl.k_cache": {"inputs": [[[12.8125]]]}, "model.layers.48.self_attn.attn.impl.v_cache": {"inputs": [[[1.359375]]]}, "model.layers.48.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.59375]], [[12.8125]], [[1.140625]]], "outputs": [[[0.40625]], [[1.0]]]}, "model.layers.48.mlp.gate_up_proj": {"inputs": [[[6.84375]]], "params": {"weight": [[0.435546875]]}}, "model.layers.48.mlp.down_proj": {"inputs": [[[3.078125]]], "outputs": [[[0.447265625]], [[2.171875]]], "params": {"weight": [[0.373046875]]}}, "model.layers.49.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.44921875]]}}, "model.layers.49.self_attn.o_proj": {"inputs": [[[0.4609375]]], "outputs": [[[0.333984375]], [[0.4140625]]], "params": {"weight": [[0.181640625]]}}, "model.layers.49.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.50390625]], [[12.3125]]]}, "model.layers.49.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.94921875]]]}, "model.layers.49.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.9375]]]}, "model.layers.49.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.40625]]]}, "model.layers.49.self_attn.attn.impl.k_cache": {"inputs": [[[12.3125]]]}, "model.layers.49.self_attn.attn.impl.v_cache": {"inputs": [[[0.94921875]]]}, "model.layers.49.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.5]], [[12.3125]], [[0.80078125]]], "outputs": [[[0.421875]], [[1.0]]]}, "model.layers.49.mlp.gate_up_proj": {"inputs": [[[6.6875]]], "params": {"weight": [[0.341796875]]}}, "model.layers.49.mlp.down_proj": {"inputs": [[[3.15625]]], "outputs": [[[1.15625]], [[3.046875]]], "params": {"weight": [[1.515625]]}}, "model.layers.50.self_attn.qkv_proj": {"inputs": [[[11.125]]], "params": {"weight": [[0.63671875]]}}, "model.layers.50.self_attn.o_proj": {"inputs": [[[0.41796875]]], "outputs": [[[0.2080078125]], [[0.4453125]]], "params": {"weight": [[0.326171875]]}}, "model.layers.50.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.47265625]], [[12.25]]]}, "model.layers.50.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.5625]]]}, "model.layers.50.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.84375]]]}, "model.layers.50.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.6875]]]}, "model.layers.50.self_attn.attn.impl.k_cache": {"inputs": [[[12.25]]]}, "model.layers.50.self_attn.attn.impl.v_cache": {"inputs": [[[4.5625]]]}, "model.layers.50.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.875]], [[12.25]], [[4.5625]]], "outputs": [[[0.326171875]], [[1.0]]]}, "model.layers.50.mlp.gate_up_proj": {"inputs": [[[6.53125]]], "params": {"weight": [[0.4296875]]}}, "model.layers.50.mlp.down_proj": {"inputs": [[[7.9375]]], "outputs": [[[0.92578125]], [[0.8125]]], "params": {"weight": [[0.31640625]]}}, "model.layers.51.self_attn.qkv_proj": {"inputs": [[[10.875]]], "params": {"weight": [[0.462890625]]}}, "model.layers.51.self_attn.o_proj": {"inputs": [[[0.61328125]]], "outputs": [[[0.396484375]], [[0.578125]]], "params": {"weight": [[0.35546875]]}}, "model.layers.51.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.87109375]], [[12.8125]]]}, "model.layers.51.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.578125]]]}, "model.layers.51.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[23.875]]]}, "model.layers.51.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.9375]]]}, "model.layers.51.self_attn.attn.impl.k_cache": {"inputs": [[[12.8125]]]}, "model.layers.51.self_attn.attn.impl.v_cache": {"inputs": [[[1.578125]]]}, "model.layers.51.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.125]], [[10.75]], [[1.3671875]]], "outputs": [[[0.61328125]], [[1.0]]]}, "model.layers.51.mlp.gate_up_proj": {"inputs": [[[6.46875]]], "params": {"weight": [[0.470703125]]}}, "model.layers.51.mlp.down_proj": {"inputs": [[[3.875]]], "outputs": [[[0.44921875]], [[2.796875]]], "params": {"weight": [[1.0234375]]}}, "model.layers.52.self_attn.qkv_proj": {"inputs": [[[12.375]]], "params": {"weight": [[0.408203125]]}}, "model.layers.52.self_attn.o_proj": {"inputs": [[[1.421875]]], "outputs": [[[0.671875]], [[0.9375]]], "params": {"weight": [[0.32421875]]}}, "model.layers.52.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.74609375]], [[14.3125]]]}, "model.layers.52.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.6015625]]]}, "model.layers.52.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[30.125]]]}, "model.layers.52.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[12.375]]]}, "model.layers.52.self_attn.attn.impl.k_cache": {"inputs": [[[14.3125]]]}, "model.layers.52.self_attn.attn.impl.v_cache": {"inputs": [[[1.6015625]]]}, "model.layers.52.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.96875]], [[13.5625]], [[1.6015625]]], "outputs": [[[1.421875]], [[1.0]]]}, "model.layers.52.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[0.984375]]}}, "model.layers.52.mlp.down_proj": {"inputs": [[[8.8125]]], "outputs": [[[0.85546875]], [[1.9453125]]], "params": {"weight": [[1.171875]]}}, "model.layers.53.self_attn.qkv_proj": {"inputs": [[[11.25]]], "params": {"weight": [[0.40625]]}}, "model.layers.53.self_attn.o_proj": {"inputs": [[[0.7734375]]], "outputs": [[[0.279296875]], [[0.63671875]]], "params": {"weight": [[0.287109375]]}}, "model.layers.53.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.65625]], [[11.9375]]]}, "model.layers.53.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.4453125]]]}, "model.layers.53.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.125]]]}, "model.layers.53.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.96875]]]}, "model.layers.53.self_attn.attn.impl.k_cache": {"inputs": [[[11.9375]]]}, "model.layers.53.self_attn.attn.impl.v_cache": {"inputs": [[[1.4453125]]]}, "model.layers.53.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.6875]], [[11.0625]], [[1.4453125]]], "outputs": [[[0.66015625]], [[1.0]]]}, "model.layers.53.mlp.gate_up_proj": {"inputs": [[[6.6875]]], "params": {"weight": [[0.41015625]]}}, "model.layers.53.mlp.down_proj": {"inputs": [[[11.375]]], "outputs": [[[0.59765625]], [[1.921875]]], "params": {"weight": [[1.4921875]]}}, "model.layers.54.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.703125]]}}, "model.layers.54.self_attn.o_proj": {"inputs": [[[0.59765625]]], "outputs": [[[0.66796875]], [[1.078125]]], "params": {"weight": [[0.1884765625]]}}, "model.layers.54.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.70703125]], [[18.75]]]}, "model.layers.54.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.0390625]]]}, "model.layers.54.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[16.75]]]}, "model.layers.54.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[11.125]]]}, "model.layers.54.self_attn.attn.impl.k_cache": {"inputs": [[[18.75]]]}, "model.layers.54.self_attn.attn.impl.v_cache": {"inputs": [[[1.0390625]]]}, "model.layers.54.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.125]], [[18.75]], [[0.859375]]], "outputs": [[[0.48828125]], [[1.0]]]}, "model.layers.54.mlp.gate_up_proj": {"inputs": [[[6.5625]]], "params": {"weight": [[0.283203125]]}}, "model.layers.54.mlp.down_proj": {"inputs": [[[7.8125]]], "outputs": [[[0.6171875]], [[1.2734375]]], "params": {"weight": [[0.408203125]]}}, "model.layers.55.self_attn.qkv_proj": {"inputs": [[[10.3125]]], "params": {"weight": [[0.4140625]]}}, "model.layers.55.self_attn.o_proj": {"inputs": [[[0.70703125]]], "outputs": [[[0.9765625]], [[1.28125]]], "params": {"weight": [[0.2236328125]]}}, "model.layers.55.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.63671875]], [[11.3125]]]}, "model.layers.55.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.609375]]]}, "model.layers.55.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[13.25]]]}, "model.layers.55.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[11.3125]]]}, "model.layers.55.self_attn.attn.impl.k_cache": {"inputs": [[[11.3125]]]}, "model.layers.55.self_attn.attn.impl.v_cache": {"inputs": [[[1.609375]]]}, "model.layers.55.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.28125]], [[9.375]], [[1.2890625]]], "outputs": [[[0.61328125]], [[1.0]]]}, "model.layers.55.mlp.gate_up_proj": {"inputs": [[[6.6875]]], "params": {"weight": [[0.330078125]]}}, "model.layers.55.mlp.down_proj": {"inputs": [[[6.5625]]], "outputs": [[[1.59375]], [[3.078125]]], "params": {"weight": [[1.453125]]}}, "model.layers.56.self_attn.qkv_proj": {"inputs": [[[13.6875]]], "params": {"weight": [[0.37109375]]}}, "model.layers.56.self_attn.o_proj": {"inputs": [[[1.21875]]], "outputs": [[[0.72265625]], [[1.0546875]]], "params": {"weight": [[0.10693359375]]}}, "model.layers.56.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.3515625]], [[13.5]]]}, "model.layers.56.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.9140625]]]}, "model.layers.56.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.75]]]}, "model.layers.56.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.8125]]]}, "model.layers.56.self_attn.attn.impl.k_cache": {"inputs": [[[13.5]]]}, "model.layers.56.self_attn.attn.impl.v_cache": {"inputs": [[[1.9140625]]]}, "model.layers.56.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[13.75]], [[13.5]], [[1.7578125]]], "outputs": [[[1.1171875]], [[1.0]]]}, "model.layers.56.mlp.gate_up_proj": {"inputs": [[[6.90625]]], "params": {"weight": [[0.404296875]]}}, "model.layers.56.mlp.down_proj": {"inputs": [[[4.0625]]], "outputs": [[[2.109375]], [[2.296875]]], "params": {"weight": [[0.388671875]]}}, "model.layers.57.self_attn.qkv_proj": {"inputs": [[[12.1875]]], "params": {"weight": [[0.421875]]}}, "model.layers.57.self_attn.o_proj": {"inputs": [[[0.62890625]]], "outputs": [[[0.306640625]], [[0.7578125]]], "params": {"weight": [[0.3671875]]}}, "model.layers.57.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.515625]], [[15.9375]]]}, "model.layers.57.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.78125]]]}, "model.layers.57.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[11.375]]]}, "model.layers.57.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.90625]]]}, "model.layers.57.self_attn.attn.impl.k_cache": {"inputs": [[[15.9375]]]}, "model.layers.57.self_attn.attn.impl.v_cache": {"inputs": [[[3.78125]]]}, "model.layers.57.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.5]], [[13.75]], [[3.78125]]], "outputs": [[[0.62890625]], [[1.0]]]}, "model.layers.57.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[0.353515625]]}}, "model.layers.57.mlp.down_proj": {"inputs": [[[7.09375]]], "outputs": [[[0.66015625]], [[1.9609375]]], "params": {"weight": [[0.52734375]]}}, "model.layers.58.self_attn.qkv_proj": {"inputs": [[[12.1875]]], "params": {"weight": [[0.494140625]]}}, "model.layers.58.self_attn.o_proj": {"inputs": [[[0.7421875]]], "outputs": [[[0.298828125]], [[0.66796875]]], "params": {"weight": [[0.55078125]]}}, "model.layers.58.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.474609375]], [[13.0]]]}, "model.layers.58.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.75]]]}, "model.layers.58.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[8.5625]]]}, "model.layers.58.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.59375]]]}, "model.layers.58.self_attn.attn.impl.k_cache": {"inputs": [[[13.0]]]}, "model.layers.58.self_attn.attn.impl.v_cache": {"inputs": [[[4.75]]]}, "model.layers.58.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.25]], [[12.0]], [[4.75]]], "outputs": [[[0.53515625]], [[1.0]]]}, "model.layers.58.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[0.427734375]]}}, "model.layers.58.mlp.down_proj": {"inputs": [[[7.21875]]], "outputs": [[[0.423828125]], [[0.734375]]], "params": {"weight": [[0.7734375]]}}, "model.layers.59.self_attn.qkv_proj": {"inputs": [[[10.8125]]], "params": {"weight": [[0.52734375]]}}, "model.layers.59.self_attn.o_proj": {"inputs": [[[0.7421875]]], "outputs": [[[0.251953125]], [[1.0625]]], "params": {"weight": [[0.251953125]]}}, "model.layers.59.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.37890625]], [[11.9375]]]}, "model.layers.59.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.9453125]]]}, "model.layers.59.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[15.4375]]]}, "model.layers.59.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.0625]]]}, "model.layers.59.self_attn.attn.impl.k_cache": {"inputs": [[[11.9375]]]}, "model.layers.59.self_attn.attn.impl.v_cache": {"inputs": [[[0.9453125]]]}, "model.layers.59.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[3.875]], [[10.5625]], [[0.9453125]]], "outputs": [[[0.7421875]], [[1.0]]]}, "model.layers.59.mlp.gate_up_proj": {"inputs": [[[7.21875]]], "params": {"weight": [[0.44140625]]}}, "model.layers.59.mlp.down_proj": {"inputs": [[[7.5625]]], "outputs": [[[0.578125]], [[7.1875]]], "params": {"weight": [[0.302734375]]}}, "model.layers.60.self_attn.qkv_proj": {"inputs": [[[13.3125]]], "params": {"weight": [[0.3359375]]}}, "model.layers.60.self_attn.o_proj": {"inputs": [[[1.0703125]]], "outputs": [[[0.484375]], [[0.58984375]]], "params": {"weight": [[0.123046875]]}}, "model.layers.60.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.28125]], [[10.9375]]]}, "model.layers.60.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.4375]]]}, "model.layers.60.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[16.625]]]}, "model.layers.60.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.375]]]}, "model.layers.60.self_attn.attn.impl.k_cache": {"inputs": [[[10.9375]]]}, "model.layers.60.self_attn.attn.impl.v_cache": {"inputs": [[[2.4375]]]}, "model.layers.60.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[12.625]], [[10.9375]], [[2.21875]]], "outputs": [[[0.87109375]], [[1.0]]]}, "model.layers.60.mlp.gate_up_proj": {"inputs": [[[7.34375]]], "params": {"weight": [[0.5546875]]}}, "model.layers.60.mlp.down_proj": {"inputs": [[[7.875]]], "outputs": [[[2.296875]], [[3.671875]]], "params": {"weight": [[0.6640625]]}}, "model.layers.61.self_attn.qkv_proj": {"inputs": [[[11.5]]], "params": {"weight": [[0.52734375]]}}, "model.layers.61.self_attn.o_proj": {"inputs": [[[0.46484375]]], "outputs": [[[0.26953125]], [[0.62109375]]], "params": {"weight": [[0.2119140625]]}}, "model.layers.61.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5]], [[15.1875]]]}, "model.layers.61.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.140625]]]}, "model.layers.61.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[14.3125]]]}, "model.layers.61.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.3125]]]}, "model.layers.61.self_attn.attn.impl.k_cache": {"inputs": [[[15.1875]]]}, "model.layers.61.self_attn.attn.impl.v_cache": {"inputs": [[[1.140625]]]}, "model.layers.61.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.125]], [[15.1875]], [[1.1640625]]], "outputs": [[[0.314453125]], [[1.0]]]}, "model.layers.61.mlp.gate_up_proj": {"inputs": [[[7.46875]]], "params": {"weight": [[0.435546875]]}}, "model.layers.61.mlp.down_proj": {"inputs": [[[5.21875]]], "outputs": [[[0.7421875]], [[2.421875]]], "params": {"weight": [[0.8515625]]}}, "model.layers.62.self_attn.qkv_proj": {"inputs": [[[11.5625]]], "params": {"weight": [[0.67578125]]}}, "model.layers.62.self_attn.o_proj": {"inputs": [[[0.59375]]], "outputs": [[[0.75]], [[0.76171875]]], "params": {"weight": [[0.462890625]]}}, "model.layers.62.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.447265625]], [[13.5]]]}, "model.layers.62.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.8671875]]]}, "model.layers.62.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[16.375]]]}, "model.layers.62.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[11.9375]]]}, "model.layers.62.self_attn.attn.impl.k_cache": {"inputs": [[[13.5]]]}, "model.layers.62.self_attn.attn.impl.v_cache": {"inputs": [[[1.8671875]]]}, "model.layers.62.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.9375]], [[13.5]], [[1.8671875]]], "outputs": [[[0.494140625]], [[1.0]]]}, "model.layers.62.mlp.gate_up_proj": {"inputs": [[[7.625]]], "params": {"weight": [[0.36328125]]}}, "model.layers.62.mlp.down_proj": {"inputs": [[[7.34375]]], "outputs": [[[1.0078125]], [[1.3359375]]], "params": {"weight": [[0.3828125]]}}, "model.layers.63.self_attn.qkv_proj": {"inputs": [[[10.1875]]], "params": {"weight": [[0.63671875]]}}, "model.layers.63.self_attn.o_proj": {"inputs": [[[0.546875]]], "outputs": [[[0.259765625]], [[1.1875]]], "params": {"weight": [[0.1455078125]]}}, "model.layers.63.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.40625]], [[14.4375]]]}, "model.layers.63.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.40625]]]}, "model.layers.63.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[20.75]]]}, "model.layers.63.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.875]]]}, "model.layers.63.self_attn.attn.impl.k_cache": {"inputs": [[[14.4375]]]}, "model.layers.63.self_attn.attn.impl.v_cache": {"inputs": [[[2.40625]]]}, "model.layers.63.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.5]], [[14.4375]], [[2.40625]]], "outputs": [[[0.427734375]], [[1.0]]]}, "model.layers.63.mlp.gate_up_proj": {"inputs": [[[7.9375]]], "params": {"weight": [[0.5546875]]}}, "model.layers.63.mlp.down_proj": {"inputs": [[[5.4375]]], "outputs": [[[0.73828125]], [[1.640625]]], "params": {"weight": [[1.0390625]]}}, "model.layers.64.self_attn.qkv_proj": {"inputs": [[[13.0]]], "params": {"weight": [[0.4140625]]}}, "model.layers.64.self_attn.o_proj": {"inputs": [[[1.234375]]], "outputs": [[[0.341796875]], [[1.5]]], "params": {"weight": [[0.158203125]]}}, "model.layers.64.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.86328125]], [[14.75]]]}, "model.layers.64.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.296875]]]}, "model.layers.64.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[20.5]]]}, "model.layers.64.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[12.1875]]]}, "model.layers.64.self_attn.attn.impl.k_cache": {"inputs": [[[14.75]]]}, "model.layers.64.self_attn.attn.impl.v_cache": {"inputs": [[[2.296875]]]}, "model.layers.64.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.375]], [[14.625]], [[1.875]]], "outputs": [[[0.8125]], [[1.0]]]}, "model.layers.64.mlp.gate_up_proj": {"inputs": [[[8.3125]]], "params": {"weight": [[0.68359375]]}}, "model.layers.64.mlp.down_proj": {"inputs": [[[6.03125]]], "outputs": [[[0.578125]], [[2.578125]]], "params": {"weight": [[0.37890625]]}}, "model.layers.65.self_attn.qkv_proj": {"inputs": [[[14.6875]]], "params": {"weight": [[0.77734375]]}}, "model.layers.65.self_attn.o_proj": {"inputs": [[[0.453125]]], "outputs": [[[1.75]], [[1.6875]]], "params": {"weight": [[0.1103515625]]}}, "model.layers.65.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.73828125]], [[14.8125]]]}, "model.layers.65.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.79296875]]]}, "model.layers.65.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.6875]]]}, "model.layers.65.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.625]]]}, "model.layers.65.self_attn.attn.impl.k_cache": {"inputs": [[[14.8125]]]}, "model.layers.65.self_attn.attn.impl.v_cache": {"inputs": [[[0.79296875]]]}, "model.layers.65.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.875]], [[14.75]], [[0.64453125]]], "outputs": [[[0.3984375]], [[1.0]]]}, "model.layers.65.mlp.gate_up_proj": {"inputs": [[[8.125]]], "params": {"weight": [[0.67578125]]}}, "model.layers.65.mlp.down_proj": {"inputs": [[[9.25]]], "outputs": [[[0.76171875]], [[2.015625]]], "params": {"weight": [[0.33984375]]}}, "model.layers.66.self_attn.qkv_proj": {"inputs": [[[11.375]]], "params": {"weight": [[0.4921875]]}}, "model.layers.66.self_attn.o_proj": {"inputs": [[[0.8125]]], "outputs": [[[1.109375]], [[1.3359375]]], "params": {"weight": [[0.64453125]]}}, "model.layers.66.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5234375]], [[9.875]]]}, "model.layers.66.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.5625]]]}, "model.layers.66.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[9.0]]]}, "model.layers.66.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.0]]]}, "model.layers.66.self_attn.attn.impl.k_cache": {"inputs": [[[9.875]]]}, "model.layers.66.self_attn.attn.impl.v_cache": {"inputs": [[[4.5625]]]}, "model.layers.66.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.40625]], [[9.875]], [[4.5625]]], "outputs": [[[0.4765625]], [[1.0]]]}, "model.layers.66.mlp.gate_up_proj": {"inputs": [[[7.78125]]], "params": {"weight": [[0.427734375]]}}, "model.layers.66.mlp.down_proj": {"inputs": [[[10.125]]], "outputs": [[[2.5625]], [[3.5]]], "params": {"weight": [[0.45703125]]}}, "model.layers.67.self_attn.qkv_proj": {"inputs": [[[12.6875]]], "params": {"weight": [[0.314453125]]}}, "model.layers.67.self_attn.o_proj": {"inputs": [[[1.1484375]]], "outputs": [[[0.62890625]], [[1.453125]]], "params": {"weight": [[0.373046875]]}}, "model.layers.67.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6875]], [[15.375]]]}, "model.layers.67.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.078125]]]}, "model.layers.67.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[8.625]]]}, "model.layers.67.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.5]]]}, "model.layers.67.self_attn.attn.impl.k_cache": {"inputs": [[[15.375]]]}, "model.layers.67.self_attn.attn.impl.v_cache": {"inputs": [[[2.078125]]]}, "model.layers.67.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.9375]], [[13.9375]], [[2.078125]]], "outputs": [[[0.671875]], [[1.0]]]}, "model.layers.67.mlp.gate_up_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.439453125]]}}, "model.layers.67.mlp.down_proj": {"inputs": [[[7.59375]]], "outputs": [[[0.68359375]], [[1.5546875]]], "params": {"weight": [[0.255859375]]}}, "model.layers.68.self_attn.qkv_proj": {"inputs": [[[13.75]]], "params": {"weight": [[0.359375]]}}, "model.layers.68.self_attn.o_proj": {"inputs": [[[0.84375]]], "outputs": [[[0.294921875]], [[2.09375]]], "params": {"weight": [[0.2373046875]]}}, "model.layers.68.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.28125]], [[18.125]]]}, "model.layers.68.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.34375]]]}, "model.layers.68.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[14.6875]]]}, "model.layers.68.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.375]]]}, "model.layers.68.self_attn.attn.impl.k_cache": {"inputs": [[[18.125]]]}, "model.layers.68.self_attn.attn.impl.v_cache": {"inputs": [[[2.34375]]]}, "model.layers.68.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[13.625]], [[17.125]], [[2.40625]]], "outputs": [[[0.8203125]], [[1.0]]]}, "model.layers.68.mlp.gate_up_proj": {"inputs": [[[9.0625]]], "params": {"weight": [[0.56640625]]}}, "model.layers.68.mlp.down_proj": {"inputs": [[[8.9375]]], "outputs": [[[1.6171875]], [[3.953125]]], "params": {"weight": [[0.390625]]}}, "model.layers.69.self_attn.qkv_proj": {"inputs": [[[12.0]]], "params": {"weight": [[0.419921875]]}}, "model.layers.69.self_attn.o_proj": {"inputs": [[[0.85546875]]], "outputs": [[[0.58203125]], [[2.109375]]], "params": {"weight": [[0.482421875]]}}, "model.layers.69.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.578125]], [[13.25]]]}, "model.layers.69.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[6.4375]]]}, "model.layers.69.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.96875]]]}, "model.layers.69.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.625]]]}, "model.layers.69.self_attn.attn.impl.k_cache": {"inputs": [[[13.25]]]}, "model.layers.69.self_attn.attn.impl.v_cache": {"inputs": [[[6.4375]]]}, "model.layers.69.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.78125]], [[12.5625]], [[6.4375]]], "outputs": [[[0.76171875]], [[1.0]]]}, "model.layers.69.mlp.gate_up_proj": {"inputs": [[[8.625]]], "params": {"weight": [[0.2578125]]}}, "model.layers.69.mlp.down_proj": {"inputs": [[[8.625]]], "outputs": [[[0.875]], [[1.1015625]]], "params": {"weight": [[0.21484375]]}}, "model.layers.70.self_attn.qkv_proj": {"inputs": [[[12.625]]], "params": {"weight": [[0.5703125]]}}, "model.layers.70.self_attn.o_proj": {"inputs": [[[1.1015625]]], "outputs": [[[0.484375]], [[2.5]]], "params": {"weight": [[0.2236328125]]}}, "model.layers.70.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.76953125]], [[14.8125]]]}, "model.layers.70.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.625]]]}, "model.layers.70.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[23.0]]]}, "model.layers.70.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[11.5625]]]}, "model.layers.70.self_attn.attn.impl.k_cache": {"inputs": [[[14.8125]]]}, "model.layers.70.self_attn.attn.impl.v_cache": {"inputs": [[[3.625]]]}, "model.layers.70.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.625]], [[14.8125]], [[3.625]]], "outputs": [[[0.765625]], [[1.0]]]}, "model.layers.70.mlp.gate_up_proj": {"inputs": [[[8.875]]], "params": {"weight": [[0.439453125]]}}, "model.layers.70.mlp.down_proj": {"inputs": [[[6.125]]], "outputs": [[[1.625]], [[4.875]]], "params": {"weight": [[0.478515625]]}}, "model.layers.71.self_attn.qkv_proj": {"inputs": [[[13.0625]]], "params": {"weight": [[0.2578125]]}}, "model.layers.71.self_attn.o_proj": {"inputs": [[[1.0]]], "outputs": [[[1.125]], [[5.34375]]], "params": {"weight": [[0.3359375]]}}, "model.layers.71.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.75]], [[14.4375]]]}, "model.layers.71.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.84375]]]}, "model.layers.71.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[13.125]]]}, "model.layers.71.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.1875]]]}, "model.layers.71.self_attn.attn.impl.k_cache": {"inputs": [[[14.4375]]]}, "model.layers.71.self_attn.attn.impl.v_cache": {"inputs": [[[2.84375]]]}, "model.layers.71.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.0625]], [[13.875]], [[1.9609375]]], "outputs": [[[0.96875]], [[1.0]]]}, "model.layers.71.mlp.gate_up_proj": {"inputs": [[[9.4375]]], "params": {"weight": [[0.48046875]]}}, "model.layers.71.mlp.down_proj": {"inputs": [[[7.625]]], "outputs": [[[1.703125]], [[3.953125]]], "params": {"weight": [[0.328125]]}}, "model.layers.72.self_attn.qkv_proj": {"inputs": [[[11.6875]]], "params": {"weight": [[0.232421875]]}}, "model.layers.72.self_attn.o_proj": {"inputs": [[[1.6953125]]], "outputs": [[[0.81640625]], [[3.484375]]], "params": {"weight": [[0.205078125]]}}, "model.layers.72.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.796875]], [[16.5]]]}, "model.layers.72.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.890625]]]}, "model.layers.72.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[17.875]]]}, "model.layers.72.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[12.875]]]}, "model.layers.72.self_attn.attn.impl.k_cache": {"inputs": [[[16.5]]]}, "model.layers.72.self_attn.attn.impl.v_cache": {"inputs": [[[2.890625]]]}, "model.layers.72.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.5625]], [[16.0]], [[1.8359375]]], "outputs": [[[0.8828125]], [[1.0]]]}, "model.layers.72.mlp.gate_up_proj": {"inputs": [[[9.75]]], "params": {"weight": [[0.365234375]]}}, "model.layers.72.mlp.down_proj": {"inputs": [[[9.125]]], "outputs": [[[1.40625]], [[2.53125]]], "params": {"weight": [[0.484375]]}}, "model.layers.73.self_attn.qkv_proj": {"inputs": [[[12.0]]], "params": {"weight": [[0.255859375]]}}, "model.layers.73.self_attn.o_proj": {"inputs": [[[1.40625]]], "outputs": [[[2.40625]], [[3.78125]]], "params": {"weight": [[0.4453125]]}}, "model.layers.73.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.9375]], [[17.5]]]}, "model.layers.73.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.84375]]]}, "model.layers.73.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[14.6875]]]}, "model.layers.73.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.4375]]]}, "model.layers.73.self_attn.attn.impl.k_cache": {"inputs": [[[17.5]]]}, "model.layers.73.self_attn.attn.impl.v_cache": {"inputs": [[[2.84375]]]}, "model.layers.73.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[10.9375]], [[16.375]], [[2.84375]]], "outputs": [[[1.234375]], [[1.0]]]}, "model.layers.73.mlp.gate_up_proj": {"inputs": [[[9.9375]]], "params": {"weight": [[0.53515625]]}}, "model.layers.73.mlp.down_proj": {"inputs": [[[13.8125]]], "outputs": [[[2.78125]], [[4.25]]], "params": {"weight": [[0.6796875]]}}, "model.layers.74.self_attn.qkv_proj": {"inputs": [[[11.625]]], "params": {"weight": [[0.4375]]}}, "model.layers.74.self_attn.o_proj": {"inputs": [[[1.765625]]], "outputs": [[[3.15625]], [[5.59375]]], "params": {"weight": [[0.369140625]]}}, "model.layers.74.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.890625]], [[16.5]]]}, "model.layers.74.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.03125]]]}, "model.layers.74.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[52.25]]]}, "model.layers.74.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[18.875]]]}, "model.layers.74.self_attn.attn.impl.k_cache": {"inputs": [[[16.5]]]}, "model.layers.74.self_attn.attn.impl.v_cache": {"inputs": [[[3.03125]]]}, "model.layers.74.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[10.8125]], [[15.6875]], [[2.640625]]], "outputs": [[[1.765625]], [[1.0]]]}, "model.layers.74.mlp.gate_up_proj": {"inputs": [[[8.8125]]], "params": {"weight": [[0.439453125]]}}, "model.layers.74.mlp.down_proj": {"inputs": [[[7.5]]], "outputs": [[[2.53125]], [[6.9375]]], "params": {"weight": [[1.1015625]]}}, "model.layers.75.self_attn.qkv_proj": {"inputs": [[[11.5625]]], "params": {"weight": [[0.384765625]]}}, "model.layers.75.self_attn.o_proj": {"inputs": [[[1.6015625]]], "outputs": [[[1.25]], [[6.65625]]], "params": {"weight": [[0.30078125]]}}, "model.layers.75.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.1328125]], [[16.25]]]}, "model.layers.75.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.109375]]]}, "model.layers.75.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[34.0]]]}, "model.layers.75.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.875]]]}, "model.layers.75.self_attn.attn.impl.k_cache": {"inputs": [[[16.25]]]}, "model.layers.75.self_attn.attn.impl.v_cache": {"inputs": [[[2.109375]]]}, "model.layers.75.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[14.125]], [[15.4375]], [[1.8671875]]], "outputs": [[[1.2578125]], [[1.0]]]}, "model.layers.75.mlp.gate_up_proj": {"inputs": [[[8.8125]]], "params": {"weight": [[0.291015625]]}}, "model.layers.75.mlp.down_proj": {"inputs": [[[12.625]]], "outputs": [[[9.875]], [[12.4375]]], "params": {"weight": [[0.953125]]}}, "model.layers.76.self_attn.qkv_proj": {"inputs": [[[9.0]]], "params": {"weight": [[0.318359375]]}}, "model.layers.76.self_attn.o_proj": {"inputs": [[[1.7109375]]], "outputs": [[[3.03125]], [[5.3125]]], "params": {"weight": [[0.3671875]]}}, "model.layers.76.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.86328125]], [[14.4375]]]}, "model.layers.76.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.40625]]]}, "model.layers.76.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[60.0]]]}, "model.layers.76.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[28.0]]]}, "model.layers.76.self_attn.attn.impl.k_cache": {"inputs": [[[14.4375]]]}, "model.layers.76.self_attn.attn.impl.v_cache": {"inputs": [[[2.40625]]]}, "model.layers.76.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[10.625]], [[12.375]], [[2.375]]], "outputs": [[[1.0078125]], [[1.0]]]}, "model.layers.76.mlp.gate_up_proj": {"inputs": [[[9.375]]], "params": {"weight": [[0.48046875]]}}, "model.layers.76.mlp.down_proj": {"inputs": [[[17.25]]], "outputs": [[[2.078125]], [[4.90625]]], "params": {"weight": [[0.70703125]]}}, "model.layers.77.self_attn.qkv_proj": {"inputs": [[[9.375]]], "params": {"weight": [[0.2041015625]]}}, "model.layers.77.self_attn.o_proj": {"inputs": [[[1.9765625]]], "outputs": [[[2.703125]], [[3.96875]]], "params": {"weight": [[0.404296875]]}}, "model.layers.77.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.87890625]], [[13.375]]]}, "model.layers.77.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.65625]]]}, "model.layers.77.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[58.5]]]}, "model.layers.77.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[22.5]]]}, "model.layers.77.self_attn.attn.impl.k_cache": {"inputs": [[[13.375]]]}, "model.layers.77.self_attn.attn.impl.v_cache": {"inputs": [[[2.65625]]]}, "model.layers.77.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[11.3125]], [[13.0]], [[1.4375]]], "outputs": [[[1.2421875]], [[1.0]]]}, "model.layers.77.mlp.gate_up_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.40234375]]}}, "model.layers.77.mlp.down_proj": {"inputs": [[[11.5]]], "outputs": [[[4.03125]], [[6.8125]]], "params": {"weight": [[0.435546875]]}}, "model.layers.78.self_attn.qkv_proj": {"inputs": [[[8.625]]], "params": {"weight": [[0.35546875]]}}, "model.layers.78.self_attn.o_proj": {"inputs": [[[1.46875]]], "outputs": [[[2.75]], [[4.03125]]], "params": {"weight": [[0.2255859375]]}}, "model.layers.78.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.86328125]], [[11.8125]]]}, "model.layers.78.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.328125]]]}, "model.layers.78.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[31.75]]]}, "model.layers.78.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[12.625]]]}, "model.layers.78.self_attn.attn.impl.k_cache": {"inputs": [[[11.8125]]]}, "model.layers.78.self_attn.attn.impl.v_cache": {"inputs": [[[2.328125]]]}, "model.layers.78.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[11.875]], [[11.3125]], [[1.9765625]]], "outputs": [[[1.15625]], [[1.0]]]}, "model.layers.78.mlp.gate_up_proj": {"inputs": [[[14.0]]], "params": {"weight": [[0.3828125]]}}, "model.layers.78.mlp.down_proj": {"inputs": [[[18.25]]], "outputs": [[[86.5]], [[125.0]]], "params": {"weight": [[0.408203125]]}}, "model.layers.79.self_attn.qkv_proj": {"inputs": [[[9.375]]], "params": {"weight": [[0.70703125]]}}, "model.layers.79.self_attn.o_proj": {"inputs": [[[1.7109375]]], "outputs": [[[4.125]], [[5.15625]]], "params": {"weight": [[0.33203125]]}}, "model.layers.79.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6953125]], [[12.0625]]]}, "model.layers.79.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.859375]]]}, "model.layers.79.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[50.25]]]}, "model.layers.79.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[22.75]]]}, "model.layers.79.self_attn.attn.impl.k_cache": {"inputs": [[[12.0625]]]}, "model.layers.79.self_attn.attn.impl.v_cache": {"inputs": [[[1.859375]]]}, "model.layers.79.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.0]], [[9.8125]], [[1.5234375]]], "outputs": [[[1.46875]], [[1.0]]]}, "model.layers.79.mlp.gate_up_proj": {"inputs": [[[17.25]]], "params": {"weight": [[0.40234375]]}}, "model.layers.79.mlp.down_proj": {"inputs": [[[66.0]]], "outputs": [[[155.0]], [[402.0]]], "params": {"weight": [[0.54296875]]}}, "lm_head": {"inputs": [[[81.0]]], "params": {"weight": [[0.279296875]]}}}}

quant/g3/inc_output_hooks_maxabs_1_4.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b37e1e61f3612dd47663627ae969cf1037eb0656e27dcc2cc99d4764f8de0bc
+size 206298

quant/g3/inc_output_hooks_maxabs_1_4_mod_list.json

@@ -0,0 +1,963 @@
+[
+    "model.layers.0.self_attn.qkv_proj",
+    "model.layers.0.self_attn.o_proj",
+    "model.layers.0.self_attn.attn.impl.matmul_qk",
+    "model.layers.0.self_attn.attn.impl.softmax",
+    "model.layers.0.self_attn.attn.impl.matmul_av",
+    "model.layers.0.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.0.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.0.self_attn.attn.impl.k_cache",
+    "model.layers.0.self_attn.attn.impl.v_cache",
+    "model.layers.0.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.0.mlp.gate_up_proj",
+    "model.layers.0.mlp.down_proj",
+    "model.layers.1.self_attn.qkv_proj",
+    "model.layers.1.self_attn.o_proj",
+    "model.layers.1.self_attn.attn.impl.matmul_qk",
+    "model.layers.1.self_attn.attn.impl.softmax",
+    "model.layers.1.self_attn.attn.impl.matmul_av",
+    "model.layers.1.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.1.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.1.self_attn.attn.impl.k_cache",
+    "model.layers.1.self_attn.attn.impl.v_cache",
+    "model.layers.1.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.1.mlp.gate_up_proj",
+    "model.layers.1.mlp.down_proj",
+    "model.layers.2.self_attn.qkv_proj",
+    "model.layers.2.self_attn.o_proj",
+    "model.layers.2.self_attn.attn.impl.matmul_qk",
+    "model.layers.2.self_attn.attn.impl.softmax",
+    "model.layers.2.self_attn.attn.impl.matmul_av",
+    "model.layers.2.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.2.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.2.self_attn.attn.impl.k_cache",
+    "model.layers.2.self_attn.attn.impl.v_cache",
+    "model.layers.2.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.2.mlp.gate_up_proj",
+    "model.layers.2.mlp.down_proj",
+    "model.layers.3.self_attn.qkv_proj",
+    "model.layers.3.self_attn.o_proj",
+    "model.layers.3.self_attn.attn.impl.matmul_qk",
+    "model.layers.3.self_attn.attn.impl.softmax",
+    "model.layers.3.self_attn.attn.impl.matmul_av",
+    "model.layers.3.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.3.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.3.self_attn.attn.impl.k_cache",
+    "model.layers.3.self_attn.attn.impl.v_cache",
+    "model.layers.3.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.3.mlp.gate_up_proj",
+    "model.layers.3.mlp.down_proj",
+    "model.layers.4.self_attn.qkv_proj",
+    "model.layers.4.self_attn.o_proj",
+    "model.layers.4.self_attn.attn.impl.matmul_qk",
+    "model.layers.4.self_attn.attn.impl.softmax",
+    "model.layers.4.self_attn.attn.impl.matmul_av",
+    "model.layers.4.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.4.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.4.self_attn.attn.impl.k_cache",
+    "model.layers.4.self_attn.attn.impl.v_cache",
+    "model.layers.4.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.4.mlp.gate_up_proj",
+    "model.layers.4.mlp.down_proj",
+    "model.layers.5.self_attn.qkv_proj",
+    "model.layers.5.self_attn.o_proj",
+    "model.layers.5.self_attn.attn.impl.matmul_qk",
+    "model.layers.5.self_attn.attn.impl.softmax",
+    "model.layers.5.self_attn.attn.impl.matmul_av",
+    "model.layers.5.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.5.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.5.self_attn.attn.impl.k_cache",
+    "model.layers.5.self_attn.attn.impl.v_cache",
+    "model.layers.5.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.5.mlp.gate_up_proj",
+    "model.layers.5.mlp.down_proj",
+    "model.layers.6.self_attn.qkv_proj",
+    "model.layers.6.self_attn.o_proj",
+    "model.layers.6.self_attn.attn.impl.matmul_qk",
+    "model.layers.6.self_attn.attn.impl.softmax",
+    "model.layers.6.self_attn.attn.impl.matmul_av",
+    "model.layers.6.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.6.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.6.self_attn.attn.impl.k_cache",
+    "model.layers.6.self_attn.attn.impl.v_cache",
+    "model.layers.6.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.6.mlp.gate_up_proj",
+    "model.layers.6.mlp.down_proj",
+    "model.layers.7.self_attn.qkv_proj",
+    "model.layers.7.self_attn.o_proj",
+    "model.layers.7.self_attn.attn.impl.matmul_qk",
+    "model.layers.7.self_attn.attn.impl.softmax",
+    "model.layers.7.self_attn.attn.impl.matmul_av",
+    "model.layers.7.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.7.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.7.self_attn.attn.impl.k_cache",
+    "model.layers.7.self_attn.attn.impl.v_cache",
+    "model.layers.7.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.7.mlp.gate_up_proj",
+    "model.layers.7.mlp.down_proj",
+    "model.layers.8.self_attn.qkv_proj",
+    "model.layers.8.self_attn.o_proj",
+    "model.layers.8.self_attn.attn.impl.matmul_qk",
+    "model.layers.8.self_attn.attn.impl.softmax",
+    "model.layers.8.self_attn.attn.impl.matmul_av",
+    "model.layers.8.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.8.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.8.self_attn.attn.impl.k_cache",
+    "model.layers.8.self_attn.attn.impl.v_cache",
+    "model.layers.8.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.8.mlp.gate_up_proj",
+    "model.layers.8.mlp.down_proj",
+    "model.layers.9.self_attn.qkv_proj",
+    "model.layers.9.self_attn.o_proj",
+    "model.layers.9.self_attn.attn.impl.matmul_qk",
+    "model.layers.9.self_attn.attn.impl.softmax",
+    "model.layers.9.self_attn.attn.impl.matmul_av",
+    "model.layers.9.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.9.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.9.self_attn.attn.impl.k_cache",
+    "model.layers.9.self_attn.attn.impl.v_cache",
+    "model.layers.9.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.9.mlp.gate_up_proj",
+    "model.layers.9.mlp.down_proj",
+    "model.layers.10.self_attn.qkv_proj",
+    "model.layers.10.self_attn.o_proj",
+    "model.layers.10.self_attn.attn.impl.matmul_qk",
+    "model.layers.10.self_attn.attn.impl.softmax",
+    "model.layers.10.self_attn.attn.impl.matmul_av",
+    "model.layers.10.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.10.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.10.self_attn.attn.impl.k_cache",
+    "model.layers.10.self_attn.attn.impl.v_cache",
+    "model.layers.10.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.10.mlp.gate_up_proj",
+    "model.layers.10.mlp.down_proj",
+    "model.layers.11.self_attn.qkv_proj",
+    "model.layers.11.self_attn.o_proj",
+    "model.layers.11.self_attn.attn.impl.matmul_qk",
+    "model.layers.11.self_attn.attn.impl.softmax",
+    "model.layers.11.self_attn.attn.impl.matmul_av",
+    "model.layers.11.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.11.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.11.self_attn.attn.impl.k_cache",
+    "model.layers.11.self_attn.attn.impl.v_cache",
+    "model.layers.11.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.11.mlp.gate_up_proj",
+    "model.layers.11.mlp.down_proj",
+    "model.layers.12.self_attn.qkv_proj",
+    "model.layers.12.self_attn.o_proj",
+    "model.layers.12.self_attn.attn.impl.matmul_qk",
+    "model.layers.12.self_attn.attn.impl.softmax",
+    "model.layers.12.self_attn.attn.impl.matmul_av",
+    "model.layers.12.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.12.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.12.self_attn.attn.impl.k_cache",
+    "model.layers.12.self_attn.attn.impl.v_cache",
+    "model.layers.12.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.12.mlp.gate_up_proj",
+    "model.layers.12.mlp.down_proj",
+    "model.layers.13.self_attn.qkv_proj",
+    "model.layers.13.self_attn.o_proj",
+    "model.layers.13.self_attn.attn.impl.matmul_qk",
+    "model.layers.13.self_attn.attn.impl.softmax",
+    "model.layers.13.self_attn.attn.impl.matmul_av",
+    "model.layers.13.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.13.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.13.self_attn.attn.impl.k_cache",
+    "model.layers.13.self_attn.attn.impl.v_cache",
+    "model.layers.13.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.13.mlp.gate_up_proj",
+    "model.layers.13.mlp.down_proj",
+    "model.layers.14.self_attn.qkv_proj",
+    "model.layers.14.self_attn.o_proj",
+    "model.layers.14.self_attn.attn.impl.matmul_qk",
+    "model.layers.14.self_attn.attn.impl.softmax",
+    "model.layers.14.self_attn.attn.impl.matmul_av",
+    "model.layers.14.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.14.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.14.self_attn.attn.impl.k_cache",
+    "model.layers.14.self_attn.attn.impl.v_cache",
+    "model.layers.14.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.14.mlp.gate_up_proj",
+    "model.layers.14.mlp.down_proj",
+    "model.layers.15.self_attn.qkv_proj",
+    "model.layers.15.self_attn.o_proj",
+    "model.layers.15.self_attn.attn.impl.matmul_qk",
+    "model.layers.15.self_attn.attn.impl.softmax",
+    "model.layers.15.self_attn.attn.impl.matmul_av",
+    "model.layers.15.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.15.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.15.self_attn.attn.impl.k_cache",
+    "model.layers.15.self_attn.attn.impl.v_cache",
+    "model.layers.15.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.15.mlp.gate_up_proj",
+    "model.layers.15.mlp.down_proj",
+    "model.layers.16.self_attn.qkv_proj",
+    "model.layers.16.self_attn.o_proj",
+    "model.layers.16.self_attn.attn.impl.matmul_qk",
+    "model.layers.16.self_attn.attn.impl.softmax",
+    "model.layers.16.self_attn.attn.impl.matmul_av",
+    "model.layers.16.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.16.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.16.self_attn.attn.impl.k_cache",
+    "model.layers.16.self_attn.attn.impl.v_cache",
+    "model.layers.16.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.16.mlp.gate_up_proj",
+    "model.layers.16.mlp.down_proj",
+    "model.layers.17.self_attn.qkv_proj",
+    "model.layers.17.self_attn.o_proj",
+    "model.layers.17.self_attn.attn.impl.matmul_qk",
+    "model.layers.17.self_attn.attn.impl.softmax",
+    "model.layers.17.self_attn.attn.impl.matmul_av",
+    "model.layers.17.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.17.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.17.self_attn.attn.impl.k_cache",
+    "model.layers.17.self_attn.attn.impl.v_cache",
+    "model.layers.17.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.17.mlp.gate_up_proj",
+    "model.layers.17.mlp.down_proj",
+    "model.layers.18.self_attn.qkv_proj",
+    "model.layers.18.self_attn.o_proj",
+    "model.layers.18.self_attn.attn.impl.matmul_qk",
+    "model.layers.18.self_attn.attn.impl.softmax",
+    "model.layers.18.self_attn.attn.impl.matmul_av",
+    "model.layers.18.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.18.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.18.self_attn.attn.impl.k_cache",
+    "model.layers.18.self_attn.attn.impl.v_cache",
+    "model.layers.18.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.18.mlp.gate_up_proj",
+    "model.layers.18.mlp.down_proj",
+    "model.layers.19.self_attn.qkv_proj",
+    "model.layers.19.self_attn.o_proj",
+    "model.layers.19.self_attn.attn.impl.matmul_qk",
+    "model.layers.19.self_attn.attn.impl.softmax",
+    "model.layers.19.self_attn.attn.impl.matmul_av",
+    "model.layers.19.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.19.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.19.self_attn.attn.impl.k_cache",
+    "model.layers.19.self_attn.attn.impl.v_cache",
+    "model.layers.19.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.19.mlp.gate_up_proj",
+    "model.layers.19.mlp.down_proj",
+    "model.layers.20.self_attn.qkv_proj",
+    "model.layers.20.self_attn.o_proj",
+    "model.layers.20.self_attn.attn.impl.matmul_qk",
+    "model.layers.20.self_attn.attn.impl.softmax",
+    "model.layers.20.self_attn.attn.impl.matmul_av",
+    "model.layers.20.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.20.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.20.self_attn.attn.impl.k_cache",
+    "model.layers.20.self_attn.attn.impl.v_cache",
+    "model.layers.20.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.20.mlp.gate_up_proj",
+    "model.layers.20.mlp.down_proj",
+    "model.layers.21.self_attn.qkv_proj",
+    "model.layers.21.self_attn.o_proj",
+    "model.layers.21.self_attn.attn.impl.matmul_qk",
+    "model.layers.21.self_attn.attn.impl.softmax",
+    "model.layers.21.self_attn.attn.impl.matmul_av",
+    "model.layers.21.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.21.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.21.self_attn.attn.impl.k_cache",
+    "model.layers.21.self_attn.attn.impl.v_cache",
+    "model.layers.21.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.21.mlp.gate_up_proj",
+    "model.layers.21.mlp.down_proj",
+    "model.layers.22.self_attn.qkv_proj",
+    "model.layers.22.self_attn.o_proj",
+    "model.layers.22.self_attn.attn.impl.matmul_qk",
+    "model.layers.22.self_attn.attn.impl.softmax",
+    "model.layers.22.self_attn.attn.impl.matmul_av",
+    "model.layers.22.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.22.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.22.self_attn.attn.impl.k_cache",
+    "model.layers.22.self_attn.attn.impl.v_cache",
+    "model.layers.22.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.22.mlp.gate_up_proj",
+    "model.layers.22.mlp.down_proj",
+    "model.layers.23.self_attn.qkv_proj",
+    "model.layers.23.self_attn.o_proj",
+    "model.layers.23.self_attn.attn.impl.matmul_qk",
+    "model.layers.23.self_attn.attn.impl.softmax",
+    "model.layers.23.self_attn.attn.impl.matmul_av",
+    "model.layers.23.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.23.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.23.self_attn.attn.impl.k_cache",
+    "model.layers.23.self_attn.attn.impl.v_cache",
+    "model.layers.23.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.23.mlp.gate_up_proj",
+    "model.layers.23.mlp.down_proj",
+    "model.layers.24.self_attn.qkv_proj",
+    "model.layers.24.self_attn.o_proj",
+    "model.layers.24.self_attn.attn.impl.matmul_qk",
+    "model.layers.24.self_attn.attn.impl.softmax",
+    "model.layers.24.self_attn.attn.impl.matmul_av",
+    "model.layers.24.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.24.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.24.self_attn.attn.impl.k_cache",
+    "model.layers.24.self_attn.attn.impl.v_cache",
+    "model.layers.24.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.24.mlp.gate_up_proj",
+    "model.layers.24.mlp.down_proj",
+    "model.layers.25.self_attn.qkv_proj",
+    "model.layers.25.self_attn.o_proj",
+    "model.layers.25.self_attn.attn.impl.matmul_qk",
+    "model.layers.25.self_attn.attn.impl.softmax",
+    "model.layers.25.self_attn.attn.impl.matmul_av",
+    "model.layers.25.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.25.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.25.self_attn.attn.impl.k_cache",
+    "model.layers.25.self_attn.attn.impl.v_cache",
+    "model.layers.25.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.25.mlp.gate_up_proj",
+    "model.layers.25.mlp.down_proj",
+    "model.layers.26.self_attn.qkv_proj",
+    "model.layers.26.self_attn.o_proj",
+    "model.layers.26.self_attn.attn.impl.matmul_qk",
+    "model.layers.26.self_attn.attn.impl.softmax",
+    "model.layers.26.self_attn.attn.impl.matmul_av",
+    "model.layers.26.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.26.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.26.self_attn.attn.impl.k_cache",
+    "model.layers.26.self_attn.attn.impl.v_cache",
+    "model.layers.26.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.26.mlp.gate_up_proj",
+    "model.layers.26.mlp.down_proj",
+    "model.layers.27.self_attn.qkv_proj",
+    "model.layers.27.self_attn.o_proj",
+    "model.layers.27.self_attn.attn.impl.matmul_qk",
+    "model.layers.27.self_attn.attn.impl.softmax",
+    "model.layers.27.self_attn.attn.impl.matmul_av",
+    "model.layers.27.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.27.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.27.self_attn.attn.impl.k_cache",
+    "model.layers.27.self_attn.attn.impl.v_cache",
+    "model.layers.27.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.27.mlp.gate_up_proj",
+    "model.layers.27.mlp.down_proj",
+    "model.layers.28.self_attn.qkv_proj",
+    "model.layers.28.self_attn.o_proj",
+    "model.layers.28.self_attn.attn.impl.matmul_qk",
+    "model.layers.28.self_attn.attn.impl.softmax",
+    "model.layers.28.self_attn.attn.impl.matmul_av",
+    "model.layers.28.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.28.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.28.self_attn.attn.impl.k_cache",
+    "model.layers.28.self_attn.attn.impl.v_cache",
+    "model.layers.28.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.28.mlp.gate_up_proj",
+    "model.layers.28.mlp.down_proj",
+    "model.layers.29.self_attn.qkv_proj",
+    "model.layers.29.self_attn.o_proj",
+    "model.layers.29.self_attn.attn.impl.matmul_qk",
+    "model.layers.29.self_attn.attn.impl.softmax",
+    "model.layers.29.self_attn.attn.impl.matmul_av",
+    "model.layers.29.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.29.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.29.self_attn.attn.impl.k_cache",
+    "model.layers.29.self_attn.attn.impl.v_cache",
+    "model.layers.29.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.29.mlp.gate_up_proj",
+    "model.layers.29.mlp.down_proj",
+    "model.layers.30.self_attn.qkv_proj",
+    "model.layers.30.self_attn.o_proj",
+    "model.layers.30.self_attn.attn.impl.matmul_qk",
+    "model.layers.30.self_attn.attn.impl.softmax",
+    "model.layers.30.self_attn.attn.impl.matmul_av",
+    "model.layers.30.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.30.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.30.self_attn.attn.impl.k_cache",
+    "model.layers.30.self_attn.attn.impl.v_cache",
+    "model.layers.30.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.30.mlp.gate_up_proj",
+    "model.layers.30.mlp.down_proj",
+    "model.layers.31.self_attn.qkv_proj",
+    "model.layers.31.self_attn.o_proj",
+    "model.layers.31.self_attn.attn.impl.matmul_qk",
+    "model.layers.31.self_attn.attn.impl.softmax",
+    "model.layers.31.self_attn.attn.impl.matmul_av",
+    "model.layers.31.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.31.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.31.self_attn.attn.impl.k_cache",
+    "model.layers.31.self_attn.attn.impl.v_cache",
+    "model.layers.31.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.31.mlp.gate_up_proj",
+    "model.layers.31.mlp.down_proj",
+    "model.layers.32.self_attn.qkv_proj",
+    "model.layers.32.self_attn.o_proj",
+    "model.layers.32.self_attn.attn.impl.matmul_qk",
+    "model.layers.32.self_attn.attn.impl.softmax",
+    "model.layers.32.self_attn.attn.impl.matmul_av",
+    "model.layers.32.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.32.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.32.self_attn.attn.impl.k_cache",
+    "model.layers.32.self_attn.attn.impl.v_cache",
+    "model.layers.32.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.32.mlp.gate_up_proj",
+    "model.layers.32.mlp.down_proj",
+    "model.layers.33.self_attn.qkv_proj",
+    "model.layers.33.self_attn.o_proj",
+    "model.layers.33.self_attn.attn.impl.matmul_qk",
+    "model.layers.33.self_attn.attn.impl.softmax",
+    "model.layers.33.self_attn.attn.impl.matmul_av",
+    "model.layers.33.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.33.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.33.self_attn.attn.impl.k_cache",
+    "model.layers.33.self_attn.attn.impl.v_cache",
+    "model.layers.33.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.33.mlp.gate_up_proj",
+    "model.layers.33.mlp.down_proj",
+    "model.layers.34.self_attn.qkv_proj",
+    "model.layers.34.self_attn.o_proj",
+    "model.layers.34.self_attn.attn.impl.matmul_qk",
+    "model.layers.34.self_attn.attn.impl.softmax",
+    "model.layers.34.self_attn.attn.impl.matmul_av",
+    "model.layers.34.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.34.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.34.self_attn.attn.impl.k_cache",
+    "model.layers.34.self_attn.attn.impl.v_cache",
+    "model.layers.34.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.34.mlp.gate_up_proj",
+    "model.layers.34.mlp.down_proj",
+    "model.layers.35.self_attn.qkv_proj",
+    "model.layers.35.self_attn.o_proj",
+    "model.layers.35.self_attn.attn.impl.matmul_qk",
+    "model.layers.35.self_attn.attn.impl.softmax",
+    "model.layers.35.self_attn.attn.impl.matmul_av",
+    "model.layers.35.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.35.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.35.self_attn.attn.impl.k_cache",
+    "model.layers.35.self_attn.attn.impl.v_cache",
+    "model.layers.35.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.35.mlp.gate_up_proj",
+    "model.layers.35.mlp.down_proj",
+    "model.layers.36.self_attn.qkv_proj",
+    "model.layers.36.self_attn.o_proj",
+    "model.layers.36.self_attn.attn.impl.matmul_qk",
+    "model.layers.36.self_attn.attn.impl.softmax",
+    "model.layers.36.self_attn.attn.impl.matmul_av",
+    "model.layers.36.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.36.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.36.self_attn.attn.impl.k_cache",
+    "model.layers.36.self_attn.attn.impl.v_cache",
+    "model.layers.36.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.36.mlp.gate_up_proj",
+    "model.layers.36.mlp.down_proj",
+    "model.layers.37.self_attn.qkv_proj",
+    "model.layers.37.self_attn.o_proj",
+    "model.layers.37.self_attn.attn.impl.matmul_qk",
+    "model.layers.37.self_attn.attn.impl.softmax",
+    "model.layers.37.self_attn.attn.impl.matmul_av",
+    "model.layers.37.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.37.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.37.self_attn.attn.impl.k_cache",
+    "model.layers.37.self_attn.attn.impl.v_cache",
+    "model.layers.37.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.37.mlp.gate_up_proj",
+    "model.layers.37.mlp.down_proj",
+    "model.layers.38.self_attn.qkv_proj",
+    "model.layers.38.self_attn.o_proj",
+    "model.layers.38.self_attn.attn.impl.matmul_qk",
+    "model.layers.38.self_attn.attn.impl.softmax",
+    "model.layers.38.self_attn.attn.impl.matmul_av",
+    "model.layers.38.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.38.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.38.self_attn.attn.impl.k_cache",
+    "model.layers.38.self_attn.attn.impl.v_cache",
+    "model.layers.38.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.38.mlp.gate_up_proj",
+    "model.layers.38.mlp.down_proj",
+    "model.layers.39.self_attn.qkv_proj",
+    "model.layers.39.self_attn.o_proj",
+    "model.layers.39.self_attn.attn.impl.matmul_qk",
+    "model.layers.39.self_attn.attn.impl.softmax",
+    "model.layers.39.self_attn.attn.impl.matmul_av",
+    "model.layers.39.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.39.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.39.self_attn.attn.impl.k_cache",
+    "model.layers.39.self_attn.attn.impl.v_cache",
+    "model.layers.39.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.39.mlp.gate_up_proj",
+    "model.layers.39.mlp.down_proj",
+    "model.layers.40.self_attn.qkv_proj",
+    "model.layers.40.self_attn.o_proj",
+    "model.layers.40.self_attn.attn.impl.matmul_qk",
+    "model.layers.40.self_attn.attn.impl.softmax",
+    "model.layers.40.self_attn.attn.impl.matmul_av",
+    "model.layers.40.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.40.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.40.self_attn.attn.impl.k_cache",
+    "model.layers.40.self_attn.attn.impl.v_cache",
+    "model.layers.40.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.40.mlp.gate_up_proj",
+    "model.layers.40.mlp.down_proj",
+    "model.layers.41.self_attn.qkv_proj",
+    "model.layers.41.self_attn.o_proj",
+    "model.layers.41.self_attn.attn.impl.matmul_qk",
+    "model.layers.41.self_attn.attn.impl.softmax",
+    "model.layers.41.self_attn.attn.impl.matmul_av",
+    "model.layers.41.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.41.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.41.self_attn.attn.impl.k_cache",
+    "model.layers.41.self_attn.attn.impl.v_cache",
+    "model.layers.41.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.41.mlp.gate_up_proj",
+    "model.layers.41.mlp.down_proj",
+    "model.layers.42.self_attn.qkv_proj",
+    "model.layers.42.self_attn.o_proj",
+    "model.layers.42.self_attn.attn.impl.matmul_qk",
+    "model.layers.42.self_attn.attn.impl.softmax",
+    "model.layers.42.self_attn.attn.impl.matmul_av",
+    "model.layers.42.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.42.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.42.self_attn.attn.impl.k_cache",
+    "model.layers.42.self_attn.attn.impl.v_cache",
+    "model.layers.42.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.42.mlp.gate_up_proj",
+    "model.layers.42.mlp.down_proj",
+    "model.layers.43.self_attn.qkv_proj",
+    "model.layers.43.self_attn.o_proj",
+    "model.layers.43.self_attn.attn.impl.matmul_qk",
+    "model.layers.43.self_attn.attn.impl.softmax",
+    "model.layers.43.self_attn.attn.impl.matmul_av",
+    "model.layers.43.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.43.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.43.self_attn.attn.impl.k_cache",
+    "model.layers.43.self_attn.attn.impl.v_cache",
+    "model.layers.43.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.43.mlp.gate_up_proj",
+    "model.layers.43.mlp.down_proj",
+    "model.layers.44.self_attn.qkv_proj",
+    "model.layers.44.self_attn.o_proj",
+    "model.layers.44.self_attn.attn.impl.matmul_qk",
+    "model.layers.44.self_attn.attn.impl.softmax",
+    "model.layers.44.self_attn.attn.impl.matmul_av",
+    "model.layers.44.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.44.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.44.self_attn.attn.impl.k_cache",
+    "model.layers.44.self_attn.attn.impl.v_cache",
+    "model.layers.44.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.44.mlp.gate_up_proj",
+    "model.layers.44.mlp.down_proj",
+    "model.layers.45.self_attn.qkv_proj",
+    "model.layers.45.self_attn.o_proj",
+    "model.layers.45.self_attn.attn.impl.matmul_qk",
+    "model.layers.45.self_attn.attn.impl.softmax",
+    "model.layers.45.self_attn.attn.impl.matmul_av",
+    "model.layers.45.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.45.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.45.self_attn.attn.impl.k_cache",
+    "model.layers.45.self_attn.attn.impl.v_cache",
+    "model.layers.45.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.45.mlp.gate_up_proj",
+    "model.layers.45.mlp.down_proj",
+    "model.layers.46.self_attn.qkv_proj",
+    "model.layers.46.self_attn.o_proj",
+    "model.layers.46.self_attn.attn.impl.matmul_qk",
+    "model.layers.46.self_attn.attn.impl.softmax",
+    "model.layers.46.self_attn.attn.impl.matmul_av",
+    "model.layers.46.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.46.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.46.self_attn.attn.impl.k_cache",
+    "model.layers.46.self_attn.attn.impl.v_cache",
+    "model.layers.46.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.46.mlp.gate_up_proj",
+    "model.layers.46.mlp.down_proj",
+    "model.layers.47.self_attn.qkv_proj",
+    "model.layers.47.self_attn.o_proj",
+    "model.layers.47.self_attn.attn.impl.matmul_qk",
+    "model.layers.47.self_attn.attn.impl.softmax",
+    "model.layers.47.self_attn.attn.impl.matmul_av",
+    "model.layers.47.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.47.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.47.self_attn.attn.impl.k_cache",
+    "model.layers.47.self_attn.attn.impl.v_cache",
+    "model.layers.47.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.47.mlp.gate_up_proj",
+    "model.layers.47.mlp.down_proj",
+    "model.layers.48.self_attn.qkv_proj",
+    "model.layers.48.self_attn.o_proj",
+    "model.layers.48.self_attn.attn.impl.matmul_qk",
+    "model.layers.48.self_attn.attn.impl.softmax",
+    "model.layers.48.self_attn.attn.impl.matmul_av",
+    "model.layers.48.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.48.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.48.self_attn.attn.impl.k_cache",
+    "model.layers.48.self_attn.attn.impl.v_cache",
+    "model.layers.48.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.48.mlp.gate_up_proj",
+    "model.layers.48.mlp.down_proj",
+    "model.layers.49.self_attn.qkv_proj",
+    "model.layers.49.self_attn.o_proj",
+    "model.layers.49.self_attn.attn.impl.matmul_qk",
+    "model.layers.49.self_attn.attn.impl.softmax",
+    "model.layers.49.self_attn.attn.impl.matmul_av",
+    "model.layers.49.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.49.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.49.self_attn.attn.impl.k_cache",
+    "model.layers.49.self_attn.attn.impl.v_cache",
+    "model.layers.49.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.49.mlp.gate_up_proj",
+    "model.layers.49.mlp.down_proj",
+    "model.layers.50.self_attn.qkv_proj",
+    "model.layers.50.self_attn.o_proj",
+    "model.layers.50.self_attn.attn.impl.matmul_qk",
+    "model.layers.50.self_attn.attn.impl.softmax",
+    "model.layers.50.self_attn.attn.impl.matmul_av",
+    "model.layers.50.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.50.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.50.self_attn.attn.impl.k_cache",
+    "model.layers.50.self_attn.attn.impl.v_cache",
+    "model.layers.50.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.50.mlp.gate_up_proj",
+    "model.layers.50.mlp.down_proj",
+    "model.layers.51.self_attn.qkv_proj",
+    "model.layers.51.self_attn.o_proj",
+    "model.layers.51.self_attn.attn.impl.matmul_qk",
+    "model.layers.51.self_attn.attn.impl.softmax",
+    "model.layers.51.self_attn.attn.impl.matmul_av",
+    "model.layers.51.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.51.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.51.self_attn.attn.impl.k_cache",
+    "model.layers.51.self_attn.attn.impl.v_cache",
+    "model.layers.51.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.51.mlp.gate_up_proj",
+    "model.layers.51.mlp.down_proj",
+    "model.layers.52.self_attn.qkv_proj",
+    "model.layers.52.self_attn.o_proj",
+    "model.layers.52.self_attn.attn.impl.matmul_qk",
+    "model.layers.52.self_attn.attn.impl.softmax",
+    "model.layers.52.self_attn.attn.impl.matmul_av",
+    "model.layers.52.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.52.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.52.self_attn.attn.impl.k_cache",
+    "model.layers.52.self_attn.attn.impl.v_cache",
+    "model.layers.52.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.52.mlp.gate_up_proj",
+    "model.layers.52.mlp.down_proj",
+    "model.layers.53.self_attn.qkv_proj",
+    "model.layers.53.self_attn.o_proj",
+    "model.layers.53.self_attn.attn.impl.matmul_qk",
+    "model.layers.53.self_attn.attn.impl.softmax",
+    "model.layers.53.self_attn.attn.impl.matmul_av",
+    "model.layers.53.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.53.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.53.self_attn.attn.impl.k_cache",
+    "model.layers.53.self_attn.attn.impl.v_cache",
+    "model.layers.53.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.53.mlp.gate_up_proj",
+    "model.layers.53.mlp.down_proj",
+    "model.layers.54.self_attn.qkv_proj",
+    "model.layers.54.self_attn.o_proj",
+    "model.layers.54.self_attn.attn.impl.matmul_qk",
+    "model.layers.54.self_attn.attn.impl.softmax",
+    "model.layers.54.self_attn.attn.impl.matmul_av",
+    "model.layers.54.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.54.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.54.self_attn.attn.impl.k_cache",
+    "model.layers.54.self_attn.attn.impl.v_cache",
+    "model.layers.54.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.54.mlp.gate_up_proj",
+    "model.layers.54.mlp.down_proj",
+    "model.layers.55.self_attn.qkv_proj",
+    "model.layers.55.self_attn.o_proj",
+    "model.layers.55.self_attn.attn.impl.matmul_qk",
+    "model.layers.55.self_attn.attn.impl.softmax",
+    "model.layers.55.self_attn.attn.impl.matmul_av",
+    "model.layers.55.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.55.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.55.self_attn.attn.impl.k_cache",
+    "model.layers.55.self_attn.attn.impl.v_cache",
+    "model.layers.55.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.55.mlp.gate_up_proj",
+    "model.layers.55.mlp.down_proj",
+    "model.layers.56.self_attn.qkv_proj",
+    "model.layers.56.self_attn.o_proj",
+    "model.layers.56.self_attn.attn.impl.matmul_qk",
+    "model.layers.56.self_attn.attn.impl.softmax",
+    "model.layers.56.self_attn.attn.impl.matmul_av",
+    "model.layers.56.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.56.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.56.self_attn.attn.impl.k_cache",
+    "model.layers.56.self_attn.attn.impl.v_cache",
+    "model.layers.56.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.56.mlp.gate_up_proj",
+    "model.layers.56.mlp.down_proj",
+    "model.layers.57.self_attn.qkv_proj",
+    "model.layers.57.self_attn.o_proj",
+    "model.layers.57.self_attn.attn.impl.matmul_qk",
+    "model.layers.57.self_attn.attn.impl.softmax",
+    "model.layers.57.self_attn.attn.impl.matmul_av",
+    "model.layers.57.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.57.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.57.self_attn.attn.impl.k_cache",
+    "model.layers.57.self_attn.attn.impl.v_cache",
+    "model.layers.57.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.57.mlp.gate_up_proj",
+    "model.layers.57.mlp.down_proj",
+    "model.layers.58.self_attn.qkv_proj",
+    "model.layers.58.self_attn.o_proj",
+    "model.layers.58.self_attn.attn.impl.matmul_qk",
+    "model.layers.58.self_attn.attn.impl.softmax",
+    "model.layers.58.self_attn.attn.impl.matmul_av",
+    "model.layers.58.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.58.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.58.self_attn.attn.impl.k_cache",
+    "model.layers.58.self_attn.attn.impl.v_cache",
+    "model.layers.58.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.58.mlp.gate_up_proj",
+    "model.layers.58.mlp.down_proj",
+    "model.layers.59.self_attn.qkv_proj",
+    "model.layers.59.self_attn.o_proj",
+    "model.layers.59.self_attn.attn.impl.matmul_qk",
+    "model.layers.59.self_attn.attn.impl.softmax",
+    "model.layers.59.self_attn.attn.impl.matmul_av",
+    "model.layers.59.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.59.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.59.self_attn.attn.impl.k_cache",
+    "model.layers.59.self_attn.attn.impl.v_cache",
+    "model.layers.59.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.59.mlp.gate_up_proj",
+    "model.layers.59.mlp.down_proj",
+    "model.layers.60.self_attn.qkv_proj",
+    "model.layers.60.self_attn.o_proj",
+    "model.layers.60.self_attn.attn.impl.matmul_qk",
+    "model.layers.60.self_attn.attn.impl.softmax",
+    "model.layers.60.self_attn.attn.impl.matmul_av",
+    "model.layers.60.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.60.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.60.self_attn.attn.impl.k_cache",
+    "model.layers.60.self_attn.attn.impl.v_cache",
+    "model.layers.60.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.60.mlp.gate_up_proj",
+    "model.layers.60.mlp.down_proj",
+    "model.layers.61.self_attn.qkv_proj",
+    "model.layers.61.self_attn.o_proj",
+    "model.layers.61.self_attn.attn.impl.matmul_qk",
+    "model.layers.61.self_attn.attn.impl.softmax",
+    "model.layers.61.self_attn.attn.impl.matmul_av",
+    "model.layers.61.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.61.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.61.self_attn.attn.impl.k_cache",
+    "model.layers.61.self_attn.attn.impl.v_cache",
+    "model.layers.61.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.61.mlp.gate_up_proj",
+    "model.layers.61.mlp.down_proj",
+    "model.layers.62.self_attn.qkv_proj",
+    "model.layers.62.self_attn.o_proj",
+    "model.layers.62.self_attn.attn.impl.matmul_qk",
+    "model.layers.62.self_attn.attn.impl.softmax",
+    "model.layers.62.self_attn.attn.impl.matmul_av",
+    "model.layers.62.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.62.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.62.self_attn.attn.impl.k_cache",
+    "model.layers.62.self_attn.attn.impl.v_cache",
+    "model.layers.62.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.62.mlp.gate_up_proj",
+    "model.layers.62.mlp.down_proj",
+    "model.layers.63.self_attn.qkv_proj",
+    "model.layers.63.self_attn.o_proj",
+    "model.layers.63.self_attn.attn.impl.matmul_qk",
+    "model.layers.63.self_attn.attn.impl.softmax",
+    "model.layers.63.self_attn.attn.impl.matmul_av",
+    "model.layers.63.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.63.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.63.self_attn.attn.impl.k_cache",
+    "model.layers.63.self_attn.attn.impl.v_cache",
+    "model.layers.63.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.63.mlp.gate_up_proj",
+    "model.layers.63.mlp.down_proj",
+    "model.layers.64.self_attn.qkv_proj",
+    "model.layers.64.self_attn.o_proj",
+    "model.layers.64.self_attn.attn.impl.matmul_qk",
+    "model.layers.64.self_attn.attn.impl.softmax",
+    "model.layers.64.self_attn.attn.impl.matmul_av",
+    "model.layers.64.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.64.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.64.self_attn.attn.impl.k_cache",
+    "model.layers.64.self_attn.attn.impl.v_cache",
+    "model.layers.64.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.64.mlp.gate_up_proj",
+    "model.layers.64.mlp.down_proj",
+    "model.layers.65.self_attn.qkv_proj",
+    "model.layers.65.self_attn.o_proj",
+    "model.layers.65.self_attn.attn.impl.matmul_qk",
+    "model.layers.65.self_attn.attn.impl.softmax",
+    "model.layers.65.self_attn.attn.impl.matmul_av",
+    "model.layers.65.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.65.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.65.self_attn.attn.impl.k_cache",
+    "model.layers.65.self_attn.attn.impl.v_cache",
+    "model.layers.65.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.65.mlp.gate_up_proj",
+    "model.layers.65.mlp.down_proj",
+    "model.layers.66.self_attn.qkv_proj",
+    "model.layers.66.self_attn.o_proj",
+    "model.layers.66.self_attn.attn.impl.matmul_qk",
+    "model.layers.66.self_attn.attn.impl.softmax",
+    "model.layers.66.self_attn.attn.impl.matmul_av",
+    "model.layers.66.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.66.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.66.self_attn.attn.impl.k_cache",
+    "model.layers.66.self_attn.attn.impl.v_cache",
+    "model.layers.66.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.66.mlp.gate_up_proj",
+    "model.layers.66.mlp.down_proj",
+    "model.layers.67.self_attn.qkv_proj",
+    "model.layers.67.self_attn.o_proj",
+    "model.layers.67.self_attn.attn.impl.matmul_qk",
+    "model.layers.67.self_attn.attn.impl.softmax",
+    "model.layers.67.self_attn.attn.impl.matmul_av",
+    "model.layers.67.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.67.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.67.self_attn.attn.impl.k_cache",
+    "model.layers.67.self_attn.attn.impl.v_cache",
+    "model.layers.67.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.67.mlp.gate_up_proj",
+    "model.layers.67.mlp.down_proj",
+    "model.layers.68.self_attn.qkv_proj",
+    "model.layers.68.self_attn.o_proj",
+    "model.layers.68.self_attn.attn.impl.matmul_qk",
+    "model.layers.68.self_attn.attn.impl.softmax",
+    "model.layers.68.self_attn.attn.impl.matmul_av",
+    "model.layers.68.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.68.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.68.self_attn.attn.impl.k_cache",
+    "model.layers.68.self_attn.attn.impl.v_cache",
+    "model.layers.68.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.68.mlp.gate_up_proj",
+    "model.layers.68.mlp.down_proj",
+    "model.layers.69.self_attn.qkv_proj",
+    "model.layers.69.self_attn.o_proj",
+    "model.layers.69.self_attn.attn.impl.matmul_qk",
+    "model.layers.69.self_attn.attn.impl.softmax",
+    "model.layers.69.self_attn.attn.impl.matmul_av",
+    "model.layers.69.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.69.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.69.self_attn.attn.impl.k_cache",
+    "model.layers.69.self_attn.attn.impl.v_cache",
+    "model.layers.69.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.69.mlp.gate_up_proj",
+    "model.layers.69.mlp.down_proj",
+    "model.layers.70.self_attn.qkv_proj",
+    "model.layers.70.self_attn.o_proj",
+    "model.layers.70.self_attn.attn.impl.matmul_qk",
+    "model.layers.70.self_attn.attn.impl.softmax",
+    "model.layers.70.self_attn.attn.impl.matmul_av",
+    "model.layers.70.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.70.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.70.self_attn.attn.impl.k_cache",
+    "model.layers.70.self_attn.attn.impl.v_cache",
+    "model.layers.70.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.70.mlp.gate_up_proj",
+    "model.layers.70.mlp.down_proj",
+    "model.layers.71.self_attn.qkv_proj",
+    "model.layers.71.self_attn.o_proj",
+    "model.layers.71.self_attn.attn.impl.matmul_qk",
+    "model.layers.71.self_attn.attn.impl.softmax",
+    "model.layers.71.self_attn.attn.impl.matmul_av",
+    "model.layers.71.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.71.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.71.self_attn.attn.impl.k_cache",
+    "model.layers.71.self_attn.attn.impl.v_cache",
+    "model.layers.71.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.71.mlp.gate_up_proj",
+    "model.layers.71.mlp.down_proj",
+    "model.layers.72.self_attn.qkv_proj",
+    "model.layers.72.self_attn.o_proj",
+    "model.layers.72.self_attn.attn.impl.matmul_qk",
+    "model.layers.72.self_attn.attn.impl.softmax",
+    "model.layers.72.self_attn.attn.impl.matmul_av",
+    "model.layers.72.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.72.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.72.self_attn.attn.impl.k_cache",
+    "model.layers.72.self_attn.attn.impl.v_cache",
+    "model.layers.72.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.72.mlp.gate_up_proj",
+    "model.layers.72.mlp.down_proj",
+    "model.layers.73.self_attn.qkv_proj",
+    "model.layers.73.self_attn.o_proj",
+    "model.layers.73.self_attn.attn.impl.matmul_qk",
+    "model.layers.73.self_attn.attn.impl.softmax",
+    "model.layers.73.self_attn.attn.impl.matmul_av",
+    "model.layers.73.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.73.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.73.self_attn.attn.impl.k_cache",
+    "model.layers.73.self_attn.attn.impl.v_cache",
+    "model.layers.73.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.73.mlp.gate_up_proj",
+    "model.layers.73.mlp.down_proj",
+    "model.layers.74.self_attn.qkv_proj",
+    "model.layers.74.self_attn.o_proj",
+    "model.layers.74.self_attn.attn.impl.matmul_qk",
+    "model.layers.74.self_attn.attn.impl.softmax",
+    "model.layers.74.self_attn.attn.impl.matmul_av",
+    "model.layers.74.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.74.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.74.self_attn.attn.impl.k_cache",
+    "model.layers.74.self_attn.attn.impl.v_cache",
+    "model.layers.74.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.74.mlp.gate_up_proj",
+    "model.layers.74.mlp.down_proj",
+    "model.layers.75.self_attn.qkv_proj",
+    "model.layers.75.self_attn.o_proj",
+    "model.layers.75.self_attn.attn.impl.matmul_qk",
+    "model.layers.75.self_attn.attn.impl.softmax",
+    "model.layers.75.self_attn.attn.impl.matmul_av",
+    "model.layers.75.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.75.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.75.self_attn.attn.impl.k_cache",
+    "model.layers.75.self_attn.attn.impl.v_cache",
+    "model.layers.75.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.75.mlp.gate_up_proj",
+    "model.layers.75.mlp.down_proj",
+    "model.layers.76.self_attn.qkv_proj",
+    "model.layers.76.self_attn.o_proj",
+    "model.layers.76.self_attn.attn.impl.matmul_qk",
+    "model.layers.76.self_attn.attn.impl.softmax",
+    "model.layers.76.self_attn.attn.impl.matmul_av",
+    "model.layers.76.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.76.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.76.self_attn.attn.impl.k_cache",
+    "model.layers.76.self_attn.attn.impl.v_cache",
+    "model.layers.76.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.76.mlp.gate_up_proj",
+    "model.layers.76.mlp.down_proj",
+    "model.layers.77.self_attn.qkv_proj",
+    "model.layers.77.self_attn.o_proj",
+    "model.layers.77.self_attn.attn.impl.matmul_qk",
+    "model.layers.77.self_attn.attn.impl.softmax",
+    "model.layers.77.self_attn.attn.impl.matmul_av",
+    "model.layers.77.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.77.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.77.self_attn.attn.impl.k_cache",
+    "model.layers.77.self_attn.attn.impl.v_cache",
+    "model.layers.77.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.77.mlp.gate_up_proj",
+    "model.layers.77.mlp.down_proj",
+    "model.layers.78.self_attn.qkv_proj",
+    "model.layers.78.self_attn.o_proj",
+    "model.layers.78.self_attn.attn.impl.matmul_qk",
+    "model.layers.78.self_attn.attn.impl.softmax",
+    "model.layers.78.self_attn.attn.impl.matmul_av",
+    "model.layers.78.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.78.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.78.self_attn.attn.impl.k_cache",
+    "model.layers.78.self_attn.attn.impl.v_cache",
+    "model.layers.78.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.78.mlp.gate_up_proj",
+    "model.layers.78.mlp.down_proj",
+    "model.layers.79.self_attn.qkv_proj",
+    "model.layers.79.self_attn.o_proj",
+    "model.layers.79.self_attn.attn.impl.matmul_qk",
+    "model.layers.79.self_attn.attn.impl.softmax",
+    "model.layers.79.self_attn.attn.impl.matmul_av",
+    "model.layers.79.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.79.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.79.self_attn.attn.impl.k_cache",
+    "model.layers.79.self_attn.attn.impl.v_cache",
+    "model.layers.79.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.79.mlp.gate_up_proj",
+    "model.layers.79.mlp.down_proj",
+    "lm_head"
+]

quant/g3/inc_output_hooks_maxabs_2_4.json

@@ -0,0 +1 @@
+{"GlobalRank": null, "LocalRank": 2, "Mode": "DynamicRange", "Nodes": {"model.layers.0.self_attn.qkv_proj": {"inputs": [[[4.09375]]], "params": {"weight": [[78.5]]}}, "model.layers.0.self_attn.o_proj": {"inputs": [[[0.1318359375]]], "outputs": [[[0.140625]], [[0.248046875]]], "params": {"weight": [[0.341796875]]}}, "model.layers.0.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.55078125]], [[5.78125]]]}, "model.layers.0.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.1396484375]]]}, "model.layers.0.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[540.0]]]}, "model.layers.0.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[121.5]]]}, "model.layers.0.self_attn.attn.impl.k_cache": {"inputs": [[[5.78125]]]}, "model.layers.0.self_attn.attn.impl.v_cache": {"inputs": [[[0.1396484375]]]}, "model.layers.0.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.46875]], [[5.78125]], [[0.1337890625]]], "outputs": [[[0.1318359375]], [[1.0]]]}, "model.layers.0.mlp.gate_up_proj": {"inputs": [[[1.453125]]], "params": {"weight": [[8.4375]]}}, "model.layers.0.mlp.down_proj": {"inputs": [[[0.220703125]]], "outputs": [[[8.625]], [[52.5]]], "params": {"weight": [[0.5625]]}}, "model.layers.1.self_attn.qkv_proj": {"inputs": [[[7.40625]]], "params": {"weight": [[12.1875]]}}, "model.layers.1.self_attn.o_proj": {"inputs": [[[0.087890625]]], "outputs": [[[0.65625]], [[0.765625]]], "params": {"weight": [[0.287109375]]}}, "model.layers.1.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.52734375]], [[17.125]]]}, "model.layers.1.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.55859375]]]}, "model.layers.1.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[3.34375]]]}, "model.layers.1.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.90625]]]}, "model.layers.1.self_attn.attn.impl.k_cache": {"inputs": [[[17.125]]]}, "model.layers.1.self_attn.attn.impl.v_cache": {"inputs": [[[0.55859375]]]}, "model.layers.1.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.25]], [[17.125]], [[0.48046875]]], "outputs": [[[0.0849609375]], [[1.0]]]}, "model.layers.1.mlp.gate_up_proj": {"inputs": [[[3.09375]]], "params": {"weight": [[0.875]]}}, "model.layers.1.mlp.down_proj": {"inputs": [[[0.314453125]]], "outputs": [[[2.171875]], [[5.0625]]], "params": {"weight": [[0.46875]]}}, "model.layers.2.self_attn.qkv_proj": {"inputs": [[[6.625]]], "params": {"weight": [[3.453125]]}}, "model.layers.2.self_attn.o_proj": {"inputs": [[[0.0927734375]]], "outputs": [[[0.044189453125]], [[0.099609375]]], "params": {"weight": [[0.12451171875]]}}, "model.layers.2.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.8828125]], [[14.375]]]}, "model.layers.2.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.353515625]]]}, "model.layers.2.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[3.640625]]]}, "model.layers.2.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.203125]]]}, "model.layers.2.self_attn.attn.impl.k_cache": {"inputs": [[[14.375]]]}, "model.layers.2.self_attn.attn.impl.v_cache": {"inputs": [[[0.353515625]]]}, "model.layers.2.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.8125]], [[13.8125]], [[0.357421875]]], "outputs": [[[0.0927734375]], [[1.0]]]}, "model.layers.2.mlp.gate_up_proj": {"inputs": [[[2.578125]]], "params": {"weight": [[1.609375]]}}, "model.layers.2.mlp.down_proj": {"inputs": [[[0.734375]]], "outputs": [[[0.345703125]], [[0.72265625]]], "params": {"weight": [[0.421875]]}}, "model.layers.3.self_attn.qkv_proj": {"inputs": [[[3.796875]]], "params": {"weight": [[7.5]]}}, "model.layers.3.self_attn.o_proj": {"inputs": [[[0.1318359375]]], "outputs": [[[0.64453125]], [[0.9453125]]], "params": {"weight": [[0.185546875]]}}, "model.layers.3.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.50390625]], [[14.1875]]]}, "model.layers.3.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.416015625]]]}, "model.layers.3.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.9375]]]}, "model.layers.3.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.71875]]]}, "model.layers.3.self_attn.attn.impl.k_cache": {"inputs": [[[14.1875]]]}, "model.layers.3.self_attn.attn.impl.v_cache": {"inputs": [[[0.416015625]]]}, "model.layers.3.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.5]], [[14.1875]], [[0.376953125]]], "outputs": [[[0.1318359375]], [[1.0]]]}, "model.layers.3.mlp.gate_up_proj": {"inputs": [[[22.625]]], "params": {"weight": [[13.9375]]}}, "model.layers.3.mlp.down_proj": {"inputs": [[[143.0]]], "outputs": [[[4.90625]], [[452.0]]], "params": {"weight": [[2.171875]]}}, "model.layers.4.self_attn.qkv_proj": {"inputs": [[[4.875]]], "params": {"weight": [[0.408203125]]}}, "model.layers.4.self_attn.o_proj": {"inputs": [[[0.10205078125]]], "outputs": [[[0.1015625]], [[0.154296875]]], "params": {"weight": [[0.3125]]}}, "model.layers.4.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.703125]], [[16.625]]]}, "model.layers.4.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.140625]]]}, "model.layers.4.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.09375]]]}, "model.layers.4.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.96875]]]}, "model.layers.4.self_attn.attn.impl.k_cache": {"inputs": [[[16.625]]]}, "model.layers.4.self_attn.attn.impl.v_cache": {"inputs": [[[1.140625]]]}, "model.layers.4.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.25]], [[16.0]], [[1.1015625]]], "outputs": [[[0.10205078125]], [[1.0]]]}, "model.layers.4.mlp.gate_up_proj": {"inputs": [[[1.8828125]]], "params": {"weight": [[1.765625]]}}, "model.layers.4.mlp.down_proj": {"inputs": [[[0.52734375]]], "outputs": [[[0.15234375]], [[0.68359375]]], "params": {"weight": [[0.51953125]]}}, "model.layers.5.self_attn.qkv_proj": {"inputs": [[[4.84375]]], "params": {"weight": [[2.125]]}}, "model.layers.5.self_attn.o_proj": {"inputs": [[[0.12890625]]], "outputs": [[[0.06640625]], [[0.10888671875]]], "params": {"weight": [[0.1494140625]]}}, "model.layers.5.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5859375]], [[17.375]]]}, "model.layers.5.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.3203125]]]}, "model.layers.5.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[3.078125]]]}, "model.layers.5.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.65625]]]}, "model.layers.5.self_attn.attn.impl.k_cache": {"inputs": [[[17.375]]]}, "model.layers.5.self_attn.attn.impl.v_cache": {"inputs": [[[1.3203125]]]}, "model.layers.5.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.03125]], [[16.25]], [[1.1953125]]], "outputs": [[[0.06982421875]], [[1.0]]]}, "model.layers.5.mlp.gate_up_proj": {"inputs": [[[3.890625]]], "params": {"weight": [[3.546875]]}}, "model.layers.5.mlp.down_proj": {"inputs": [[[0.478515625]]], "outputs": [[[1.0078125]], [[7.375]]], "params": {"weight": [[0.32421875]]}}, "model.layers.6.self_attn.qkv_proj": {"inputs": [[[5.09375]]], "params": {"weight": [[0.376953125]]}}, "model.layers.6.self_attn.o_proj": {"inputs": [[[0.359375]]], "outputs": [[[0.2451171875]], [[0.279296875]]], "params": {"weight": [[0.224609375]]}}, "model.layers.6.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.50390625]], [[14.0]]]}, "model.layers.6.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.3046875]]]}, "model.layers.6.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.5]]]}, "model.layers.6.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.796875]]]}, "model.layers.6.self_attn.attn.impl.k_cache": {"inputs": [[[14.0]]]}, "model.layers.6.self_attn.attn.impl.v_cache": {"inputs": [[[1.3046875]]]}, "model.layers.6.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.21875]], [[13.375]], [[1.3046875]]], "outputs": [[[0.11669921875]], [[1.0]]]}, "model.layers.6.mlp.gate_up_proj": {"inputs": [[[3.328125]]], "params": {"weight": [[1.5859375]]}}, "model.layers.6.mlp.down_proj": {"inputs": [[[0.75390625]]], "outputs": [[[0.1982421875]], [[0.431640625]]], "params": {"weight": [[0.26171875]]}}, "model.layers.7.self_attn.qkv_proj": {"inputs": [[[5.625]]], "params": {"weight": [[0.26953125]]}}, "model.layers.7.self_attn.o_proj": {"inputs": [[[0.427734375]]], "outputs": [[[0.0849609375]], [[0.1240234375]]], "params": {"weight": [[0.1376953125]]}}, "model.layers.7.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.59765625]], [[11.25]]]}, "model.layers.7.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.8046875]]]}, "model.layers.7.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[10.375]]]}, "model.layers.7.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.90625]]]}, "model.layers.7.self_attn.attn.impl.k_cache": {"inputs": [[[11.25]]]}, "model.layers.7.self_attn.attn.impl.v_cache": {"inputs": [[[1.8046875]]]}, "model.layers.7.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.875]], [[10.125]], [[1.6015625]]], "outputs": [[[0.427734375]], [[1.0]]]}, "model.layers.7.mlp.gate_up_proj": {"inputs": [[[3.3125]]], "params": {"weight": [[0.447265625]]}}, "model.layers.7.mlp.down_proj": {"inputs": [[[0.48828125]]], "outputs": [[[0.59765625]], [[1.59375]]], "params": {"weight": [[0.259765625]]}}, "model.layers.8.self_attn.qkv_proj": {"inputs": [[[4.8125]]], "params": {"weight": [[0.25]]}}, "model.layers.8.self_attn.o_proj": {"inputs": [[[0.224609375]]], "outputs": [[[0.091796875]], [[0.138671875]]], "params": {"weight": [[0.2490234375]]}}, "model.layers.8.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.58203125]], [[22.375]]]}, "model.layers.8.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.80078125]]]}, "model.layers.8.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[3.984375]]]}, "model.layers.8.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.640625]]]}, "model.layers.8.self_attn.attn.impl.k_cache": {"inputs": [[[22.375]]]}, "model.layers.8.self_attn.attn.impl.v_cache": {"inputs": [[[0.80078125]]]}, "model.layers.8.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.03125]], [[20.75]], [[0.68359375]]], "outputs": [[[0.224609375]], [[1.0]]]}, "model.layers.8.mlp.gate_up_proj": {"inputs": [[[3.078125]]], "params": {"weight": [[1.5234375]]}}, "model.layers.8.mlp.down_proj": {"inputs": [[[0.5234375]]], "outputs": [[[0.126953125]], [[0.474609375]]], "params": {"weight": [[0.2314453125]]}}, "model.layers.9.self_attn.qkv_proj": {"inputs": [[[5.59375]]], "params": {"weight": [[0.68359375]]}}, "model.layers.9.self_attn.o_proj": {"inputs": [[[0.150390625]]], "outputs": [[[0.0517578125]], [[0.126953125]]], "params": {"weight": [[0.15234375]]}}, "model.layers.9.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5234375]], [[15.875]]]}, "model.layers.9.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.609375]]]}, "model.layers.9.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[5.375]]]}, "model.layers.9.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.21875]]]}, "model.layers.9.self_attn.attn.impl.k_cache": {"inputs": [[[15.875]]]}, "model.layers.9.self_attn.attn.impl.v_cache": {"inputs": [[[0.609375]]]}, "model.layers.9.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.65625]], [[15.25]], [[0.58203125]]], "outputs": [[[0.150390625]], [[1.0]]]}, "model.layers.9.mlp.gate_up_proj": {"inputs": [[[4.28125]]], "params": {"weight": [[7.71875]]}}, "model.layers.9.mlp.down_proj": {"inputs": [[[6.84375]]], "outputs": [[[1.4375]], [[7.15625]]], "params": {"weight": [[0.84375]]}}, "model.layers.10.self_attn.qkv_proj": {"inputs": [[[6.4375]]], "params": {"weight": [[0.328125]]}}, "model.layers.10.self_attn.o_proj": {"inputs": [[[0.294921875]]], "outputs": [[[0.11181640625]], [[0.234375]]], "params": {"weight": [[0.1416015625]]}}, "model.layers.10.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.62890625]], [[11.875]]]}, "model.layers.10.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.91015625]]]}, "model.layers.10.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.0625]]]}, "model.layers.10.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.125]]]}, "model.layers.10.self_attn.attn.impl.k_cache": {"inputs": [[[11.875]]]}, "model.layers.10.self_attn.attn.impl.v_cache": {"inputs": [[[0.91015625]]]}, "model.layers.10.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.21875]], [[10.8125]], [[0.9140625]]], "outputs": [[[0.169921875]], [[1.0]]]}, "model.layers.10.mlp.gate_up_proj": {"inputs": [[[3.125]]], "params": {"weight": [[0.326171875]]}}, "model.layers.10.mlp.down_proj": {"inputs": [[[0.76953125]]], "outputs": [[[0.10888671875]], [[0.40625]]], "params": {"weight": [[0.326171875]]}}, "model.layers.11.self_attn.qkv_proj": {"inputs": [[[7.4375]]], "params": {"weight": [[0.373046875]]}}, "model.layers.11.self_attn.o_proj": {"inputs": [[[0.19140625]]], "outputs": [[[0.08154296875]], [[0.166015625]]], "params": {"weight": [[0.19921875]]}}, "model.layers.11.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.474609375]], [[10.1875]]]}, "model.layers.11.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.046875]]]}, "model.layers.11.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[6.53125]]]}, "model.layers.11.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.890625]]]}, "model.layers.11.self_attn.attn.impl.k_cache": {"inputs": [[[10.1875]]]}, "model.layers.11.self_attn.attn.impl.v_cache": {"inputs": [[[1.046875]]]}, "model.layers.11.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.8125]], [[8.8125]], [[1.046875]]], "outputs": [[[0.19140625]], [[1.0]]]}, "model.layers.11.mlp.gate_up_proj": {"inputs": [[[3.34375]]], "params": {"weight": [[1.0390625]]}}, "model.layers.11.mlp.down_proj": {"inputs": [[[0.87890625]]], "outputs": [[[0.169921875]], [[0.75390625]]], "params": {"weight": [[0.154296875]]}}, "model.layers.12.self_attn.qkv_proj": {"inputs": [[[6.1875]]], "params": {"weight": [[0.216796875]]}}, "model.layers.12.self_attn.o_proj": {"inputs": [[[0.1591796875]]], "outputs": [[[0.1376953125]], [[0.32421875]]], "params": {"weight": [[0.1552734375]]}}, "model.layers.12.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5078125]], [[13.625]]]}, "model.layers.12.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.6796875]]]}, "model.layers.12.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.40625]]]}, "model.layers.12.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.15625]]]}, "model.layers.12.self_attn.attn.impl.k_cache": {"inputs": [[[13.625]]]}, "model.layers.12.self_attn.attn.impl.v_cache": {"inputs": [[[1.6796875]]]}, "model.layers.12.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.3125]], [[13.0]], [[1.6796875]]], "outputs": [[[0.1552734375]], [[1.0]]]}, "model.layers.12.mlp.gate_up_proj": {"inputs": [[[3.703125]]], "params": {"weight": [[0.478515625]]}}, "model.layers.12.mlp.down_proj": {"inputs": [[[0.62890625]]], "outputs": [[[0.5234375]], [[1.734375]]], "params": {"weight": [[0.25390625]]}}, "model.layers.13.self_attn.qkv_proj": {"inputs": [[[6.0]]], "params": {"weight": [[0.2041015625]]}}, "model.layers.13.self_attn.o_proj": {"inputs": [[[0.55859375]]], "outputs": [[[0.150390625]], [[0.34375]]], "params": {"weight": [[0.1552734375]]}}, "model.layers.13.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.63671875]], [[18.0]]]}, "model.layers.13.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.6875]]]}, "model.layers.13.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.90625]]]}, "model.layers.13.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.40625]]]}, "model.layers.13.self_attn.attn.impl.k_cache": {"inputs": [[[18.0]]]}, "model.layers.13.self_attn.attn.impl.v_cache": {"inputs": [[[1.6875]]]}, "model.layers.13.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.90625]], [[17.625]], [[1.484375]]], "outputs": [[[0.3359375]], [[1.0]]]}, "model.layers.13.mlp.gate_up_proj": {"inputs": [[[3.8125]]], "params": {"weight": [[0.39453125]]}}, "model.layers.13.mlp.down_proj": {"inputs": [[[1.5390625]]], "outputs": [[[0.328125]], [[0.89453125]]], "params": {"weight": [[0.265625]]}}, "model.layers.14.self_attn.qkv_proj": {"inputs": [[[6.4375]]], "params": {"weight": [[0.275390625]]}}, "model.layers.14.self_attn.o_proj": {"inputs": [[[0.578125]]], "outputs": [[[0.27734375]], [[0.490234375]]], "params": {"weight": [[0.1923828125]]}}, "model.layers.14.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.68359375]], [[12.3125]]]}, "model.layers.14.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.3203125]]]}, "model.layers.14.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[17.125]]]}, "model.layers.14.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.375]]]}, "model.layers.14.self_attn.attn.impl.k_cache": {"inputs": [[[12.3125]]]}, "model.layers.14.self_attn.attn.impl.v_cache": {"inputs": [[[1.3203125]]]}, "model.layers.14.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.65625]], [[11.5625]], [[1.0859375]]], "outputs": [[[0.447265625]], [[1.0]]]}, "model.layers.14.mlp.gate_up_proj": {"inputs": [[[3.875]]], "params": {"weight": [[0.45703125]]}}, "model.layers.14.mlp.down_proj": {"inputs": [[[0.859375]]], "outputs": [[[0.306640625]], [[0.87109375]]], "params": {"weight": [[0.2373046875]]}}, "model.layers.15.self_attn.qkv_proj": {"inputs": [[[7.0]]], "params": {"weight": [[0.2490234375]]}}, "model.layers.15.self_attn.o_proj": {"inputs": [[[0.369140625]]], "outputs": [[[0.1884765625]], [[0.4140625]]], "params": {"weight": [[0.091796875]]}}, "model.layers.15.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6171875]], [[9.875]]]}, "model.layers.15.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.5703125]]]}, "model.layers.15.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[14.375]]]}, "model.layers.15.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.25]]]}, "model.layers.15.self_attn.attn.impl.k_cache": {"inputs": [[[9.875]]]}, "model.layers.15.self_attn.attn.impl.v_cache": {"inputs": [[[1.5703125]]]}, "model.layers.15.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.8125]], [[9.75]], [[1.5703125]]], "outputs": [[[0.255859375]], [[1.0]]]}, "model.layers.15.mlp.gate_up_proj": {"inputs": [[[3.453125]]], "params": {"weight": [[0.298828125]]}}, "model.layers.15.mlp.down_proj": {"inputs": [[[1.0546875]]], "outputs": [[[0.11328125]], [[0.375]]], "params": {"weight": [[0.24609375]]}}, "model.layers.16.self_attn.qkv_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.328125]]}}, "model.layers.16.self_attn.o_proj": {"inputs": [[[0.44921875]]], "outputs": [[[0.263671875]], [[0.5234375]]], "params": {"weight": [[0.1455078125]]}}, "model.layers.16.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.8125]], [[14.375]]]}, "model.layers.16.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.5390625]]]}, "model.layers.16.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[17.375]]]}, "model.layers.16.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.5]]]}, "model.layers.16.self_attn.attn.impl.k_cache": {"inputs": [[[14.375]]]}, "model.layers.16.self_attn.attn.impl.v_cache": {"inputs": [[[1.5390625]]]}, "model.layers.16.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.15625]], [[13.0]], [[1.5390625]]], "outputs": [[[0.25]], [[1.0]]]}, "model.layers.16.mlp.gate_up_proj": {"inputs": [[[4.1875]]], "params": {"weight": [[0.2734375]]}}, "model.layers.16.mlp.down_proj": {"inputs": [[[1.0625]]], "outputs": [[[0.32421875]], [[0.6328125]]], "params": {"weight": [[0.375]]}}, "model.layers.17.self_attn.qkv_proj": {"inputs": [[[10.9375]]], "params": {"weight": [[0.1962890625]]}}, "model.layers.17.self_attn.o_proj": {"inputs": [[[0.609375]]], "outputs": [[[0.236328125]], [[0.75]]], "params": {"weight": [[0.212890625]]}}, "model.layers.17.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.796875]], [[10.375]]]}, "model.layers.17.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.234375]]]}, "model.layers.17.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[9.8125]]]}, "model.layers.17.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.5]]]}, "model.layers.17.self_attn.attn.impl.k_cache": {"inputs": [[[10.375]]]}, "model.layers.17.self_attn.attn.impl.v_cache": {"inputs": [[[1.234375]]]}, "model.layers.17.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.375]], [[10.1875]], [[1.234375]]], "outputs": [[[0.50390625]], [[1.0]]]}, "model.layers.17.mlp.gate_up_proj": {"inputs": [[[4.03125]]], "params": {"weight": [[0.416015625]]}}, "model.layers.17.mlp.down_proj": {"inputs": [[[2.03125]]], "outputs": [[[0.2353515625]], [[0.84765625]]], "params": {"weight": [[0.33984375]]}}, "model.layers.18.self_attn.qkv_proj": {"inputs": [[[11.25]]], "params": {"weight": [[0.255859375]]}}, "model.layers.18.self_attn.o_proj": {"inputs": [[[0.7421875]]], "outputs": [[[0.20703125]], [[0.640625]]], "params": {"weight": [[0.162109375]]}}, "model.layers.18.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.7734375]], [[14.4375]]]}, "model.layers.18.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.8984375]]]}, "model.layers.18.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[25.25]]]}, "model.layers.18.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.375]]]}, "model.layers.18.self_attn.attn.impl.k_cache": {"inputs": [[[14.4375]]]}, "model.layers.18.self_attn.attn.impl.v_cache": {"inputs": [[[1.8984375]]]}, "model.layers.18.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.875]], [[13.625]], [[1.0703125]]], "outputs": [[[0.455078125]], [[1.0]]]}, "model.layers.18.mlp.gate_up_proj": {"inputs": [[[7.5625]]], "params": {"weight": [[0.35546875]]}}, "model.layers.18.mlp.down_proj": {"inputs": [[[2.421875]]], "outputs": [[[6.625]], [[30.625]]], "params": {"weight": [[0.490234375]]}}, "model.layers.19.self_attn.qkv_proj": {"inputs": [[[10.6875]]], "params": {"weight": [[0.3515625]]}}, "model.layers.19.self_attn.o_proj": {"inputs": [[[0.5234375]]], "outputs": [[[0.28515625]], [[0.76171875]]], "params": {"weight": [[0.1416015625]]}}, "model.layers.19.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6640625]], [[16.625]]]}, "model.layers.19.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.1796875]]]}, "model.layers.19.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[43.5]]]}, "model.layers.19.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[17.375]]]}, "model.layers.19.self_attn.attn.impl.k_cache": {"inputs": [[[16.625]]]}, "model.layers.19.self_attn.attn.impl.v_cache": {"inputs": [[[1.1796875]]]}, "model.layers.19.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.5]], [[15.75]], [[1.046875]]], "outputs": [[[0.494140625]], [[1.0]]]}, "model.layers.19.mlp.gate_up_proj": {"inputs": [[[4.09375]]], "params": {"weight": [[0.2734375]]}}, "model.layers.19.mlp.down_proj": {"inputs": [[[3.453125]]], "outputs": [[[0.44921875]], [[1.2265625]]], "params": {"weight": [[0.86328125]]}}, "model.layers.20.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.2470703125]]}}, "model.layers.20.self_attn.o_proj": {"inputs": [[[0.5859375]]], "outputs": [[[0.2451171875]], [[1.1484375]]], "params": {"weight": [[0.2412109375]]}}, "model.layers.20.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.609375]], [[11.0]]]}, "model.layers.20.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.3203125]]]}, "model.layers.20.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[18.5]]]}, "model.layers.20.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.8125]]]}, "model.layers.20.self_attn.attn.impl.k_cache": {"inputs": [[[11.0]]]}, "model.layers.20.self_attn.attn.impl.v_cache": {"inputs": [[[1.3203125]]]}, "model.layers.20.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.21875]], [[10.375]], [[0.87109375]]], "outputs": [[[0.5859375]], [[1.0]]]}, "model.layers.20.mlp.gate_up_proj": {"inputs": [[[3.578125]]], "params": {"weight": [[0.32421875]]}}, "model.layers.20.mlp.down_proj": {"inputs": [[[2.375]]], "outputs": [[[0.73046875]], [[4.6875]]], "params": {"weight": [[0.4375]]}}, "model.layers.21.self_attn.qkv_proj": {"inputs": [[[11.4375]]], "params": {"weight": [[0.294921875]]}}, "model.layers.21.self_attn.o_proj": {"inputs": [[[0.68359375]]], "outputs": [[[0.2890625]], [[1.3984375]]], "params": {"weight": [[0.36328125]]}}, "model.layers.21.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.734375]], [[14.25]]]}, "model.layers.21.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.2890625]]]}, "model.layers.21.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[23.75]]]}, "model.layers.21.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.8125]]]}, "model.layers.21.self_attn.attn.impl.k_cache": {"inputs": [[[14.25]]]}, "model.layers.21.self_attn.attn.impl.v_cache": {"inputs": [[[1.2890625]]]}, "model.layers.21.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.125]], [[12.375]], [[1.2890625]]], "outputs": [[[0.58203125]], [[1.0]]]}, "model.layers.21.mlp.gate_up_proj": {"inputs": [[[3.578125]]], "params": {"weight": [[0.54296875]]}}, "model.layers.21.mlp.down_proj": {"inputs": [[[2.640625]]], "outputs": [[[0.68359375]], [[5.0]]], "params": {"weight": [[0.6015625]]}}, "model.layers.22.self_attn.qkv_proj": {"inputs": [[[11.0]]], "params": {"weight": [[0.26171875]]}}, "model.layers.22.self_attn.o_proj": {"inputs": [[[0.99609375]]], "outputs": [[[0.37109375]], [[1.25]]], "params": {"weight": [[0.388671875]]}}, "model.layers.22.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.89453125]], [[11.375]]]}, "model.layers.22.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[6.0]]]}, "model.layers.22.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[25.5]]]}, "model.layers.22.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.125]]]}, "model.layers.22.self_attn.attn.impl.k_cache": {"inputs": [[[11.375]]]}, "model.layers.22.self_attn.attn.impl.v_cache": {"inputs": [[[6.0]]]}, "model.layers.22.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.0]], [[10.625]], [[6.0]]], "outputs": [[[0.703125]], [[1.0]]]}, "model.layers.22.mlp.gate_up_proj": {"inputs": [[[3.25]]], "params": {"weight": [[0.3046875]]}}, "model.layers.22.mlp.down_proj": {"inputs": [[[1.71875]]], "outputs": [[[0.478515625]], [[2.578125]]], "params": {"weight": [[0.625]]}}, "model.layers.23.self_attn.qkv_proj": {"inputs": [[[11.6875]]], "params": {"weight": [[0.37109375]]}}, "model.layers.23.self_attn.o_proj": {"inputs": [[[0.6171875]]], "outputs": [[[0.578125]], [[2.140625]]], "params": {"weight": [[0.35546875]]}}, "model.layers.23.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.74609375]], [[18.125]]]}, "model.layers.23.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.234375]]]}, "model.layers.23.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[46.5]]]}, "model.layers.23.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[21.75]]]}, "model.layers.23.self_attn.attn.impl.k_cache": {"inputs": [[[18.125]]]}, "model.layers.23.self_attn.attn.impl.v_cache": {"inputs": [[[1.234375]]]}, "model.layers.23.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.40625]], [[17.75]], [[1.234375]]], "outputs": [[[0.451171875]], [[1.0]]]}, "model.layers.23.mlp.gate_up_proj": {"inputs": [[[3.265625]]], "params": {"weight": [[0.279296875]]}}, "model.layers.23.mlp.down_proj": {"inputs": [[[2.15625]]], "outputs": [[[0.5625]], [[2.75]]], "params": {"weight": [[0.6640625]]}}, "model.layers.24.self_attn.qkv_proj": {"inputs": [[[9.3125]]], "params": {"weight": [[0.453125]]}}, "model.layers.24.self_attn.o_proj": {"inputs": [[[0.5390625]]], "outputs": [[[0.2158203125]], [[2.0]]], "params": {"weight": [[0.49609375]]}}, "model.layers.24.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.67578125]], [[12.75]]]}, "model.layers.24.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.4375]]]}, "model.layers.24.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[21.25]]]}, "model.layers.24.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[11.0625]]]}, "model.layers.24.self_attn.attn.impl.k_cache": {"inputs": [[[12.75]]]}, "model.layers.24.self_attn.attn.impl.v_cache": {"inputs": [[[2.4375]]]}, "model.layers.24.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.125]], [[11.4375]], [[1.46875]]], "outputs": [[[0.41796875]], [[1.0]]]}, "model.layers.24.mlp.gate_up_proj": {"inputs": [[[3.4375]]], "params": {"weight": [[0.353515625]]}}, "model.layers.24.mlp.down_proj": {"inputs": [[[12.5625]]], "outputs": [[[2.078125]], [[14.9375]]], "params": {"weight": [[0.5234375]]}}, "model.layers.25.self_attn.qkv_proj": {"inputs": [[[10.5625]]], "params": {"weight": [[0.255859375]]}}, "model.layers.25.self_attn.o_proj": {"inputs": [[[0.5546875]]], "outputs": [[[0.375]], [[2.28125]]], "params": {"weight": [[0.77734375]]}}, "model.layers.25.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.82421875]], [[15.9375]]]}, "model.layers.25.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[7.53125]]]}, "model.layers.25.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[22.0]]]}, "model.layers.25.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.8125]]]}, "model.layers.25.self_attn.attn.impl.k_cache": {"inputs": [[[15.9375]]]}, "model.layers.25.self_attn.attn.impl.v_cache": {"inputs": [[[7.53125]]]}, "model.layers.25.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.875]], [[15.75]], [[7.53125]]], "outputs": [[[0.388671875]], [[1.0]]]}, "model.layers.25.mlp.gate_up_proj": {"inputs": [[[3.5]]], "params": {"weight": [[0.279296875]]}}, "model.layers.25.mlp.down_proj": {"inputs": [[[4.15625]]], "outputs": [[[0.9375]], [[6.28125]]], "params": {"weight": [[0.47265625]]}}, "model.layers.26.self_attn.qkv_proj": {"inputs": [[[10.875]]], "params": {"weight": [[0.30859375]]}}, "model.layers.26.self_attn.o_proj": {"inputs": [[[0.7109375]]], "outputs": [[[0.341796875]], [[2.1875]]], "params": {"weight": [[0.4140625]]}}, "model.layers.26.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.90625]], [[14.875]]]}, "model.layers.26.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.84375]]]}, "model.layers.26.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[23.25]]]}, "model.layers.26.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[12.5]]]}, "model.layers.26.self_attn.attn.impl.k_cache": {"inputs": [[[14.875]]]}, "model.layers.26.self_attn.attn.impl.v_cache": {"inputs": [[[2.84375]]]}, "model.layers.26.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.125]], [[13.9375]], [[2.84375]]], "outputs": [[[0.482421875]], [[1.0]]]}, "model.layers.26.mlp.gate_up_proj": {"inputs": [[[3.890625]]], "params": {"weight": [[0.28515625]]}}, "model.layers.26.mlp.down_proj": {"inputs": [[[1.7578125]]], "outputs": [[[0.984375]], [[3.546875]]], "params": {"weight": [[0.9375]]}}, "model.layers.27.self_attn.qkv_proj": {"inputs": [[[11.0]]], "params": {"weight": [[0.373046875]]}}, "model.layers.27.self_attn.o_proj": {"inputs": [[[0.76953125]]], "outputs": [[[0.9921875]], [[2.34375]]], "params": {"weight": [[0.255859375]]}}, "model.layers.27.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.8046875]], [[17.875]]]}, "model.layers.27.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.0390625]]]}, "model.layers.27.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[40.25]]]}, "model.layers.27.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[22.0]]]}, "model.layers.27.self_attn.attn.impl.k_cache": {"inputs": [[[17.875]]]}, "model.layers.27.self_attn.attn.impl.v_cache": {"inputs": [[[1.0390625]]]}, "model.layers.27.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.90625]], [[16.0]], [[1.0390625]]], "outputs": [[[0.76953125]], [[1.0]]]}, "model.layers.27.mlp.gate_up_proj": {"inputs": [[[4.0625]]], "params": {"weight": [[0.458984375]]}}, "model.layers.27.mlp.down_proj": {"inputs": [[[2.546875]]], "outputs": [[[0.85546875]], [[2.375]]], "params": {"weight": [[0.609375]]}}, "model.layers.28.self_attn.qkv_proj": {"inputs": [[[11.0]]], "params": {"weight": [[0.302734375]]}}, "model.layers.28.self_attn.o_proj": {"inputs": [[[0.75]]], "outputs": [[[0.82421875]], [[2.625]]], "params": {"weight": [[0.337890625]]}}, "model.layers.28.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.91796875]], [[16.0]]]}, "model.layers.28.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.359375]]]}, "model.layers.28.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[56.75]]]}, "model.layers.28.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[22.0]]]}, "model.layers.28.self_attn.attn.impl.k_cache": {"inputs": [[[16.0]]]}, "model.layers.28.self_attn.attn.impl.v_cache": {"inputs": [[[2.359375]]]}, "model.layers.28.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.5625]], [[15.375]], [[2.359375]]], "outputs": [[[0.71484375]], [[1.0]]]}, "model.layers.28.mlp.gate_up_proj": {"inputs": [[[4.78125]]], "params": {"weight": [[1.0859375]]}}, "model.layers.28.mlp.down_proj": {"inputs": [[[4.21875]]], "outputs": [[[0.8984375]], [[2.90625]]], "params": {"weight": [[0.84375]]}}, "model.layers.29.self_attn.qkv_proj": {"inputs": [[[11.125]]], "params": {"weight": [[0.26953125]]}}, "model.layers.29.self_attn.o_proj": {"inputs": [[[0.416015625]]], "outputs": [[[2.171875]], [[2.625]]], "params": {"weight": [[0.63671875]]}}, "model.layers.29.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.609375]], [[14.5]]]}, "model.layers.29.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.65625]]]}, "model.layers.29.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[13.0]]]}, "model.layers.29.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.875]]]}, "model.layers.29.self_attn.attn.impl.k_cache": {"inputs": [[[14.5]]]}, "model.layers.29.self_attn.attn.impl.v_cache": {"inputs": [[[4.65625]]]}, "model.layers.29.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.34375]], [[13.5625]], [[4.65625]]], "outputs": [[[0.416015625]], [[1.0]]]}, "model.layers.29.mlp.gate_up_proj": {"inputs": [[[4.9375]]], "params": {"weight": [[0.310546875]]}}, "model.layers.29.mlp.down_proj": {"inputs": [[[1.984375]]], "outputs": [[[1.046875]], [[2.390625]]], "params": {"weight": [[0.578125]]}}, "model.layers.30.self_attn.qkv_proj": {"inputs": [[[10.25]]], "params": {"weight": [[0.423828125]]}}, "model.layers.30.self_attn.o_proj": {"inputs": [[[0.6328125]]], "outputs": [[[0.38671875]], [[2.03125]]], "params": {"weight": [[0.26953125]]}}, "model.layers.30.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.75]], [[16.875]]]}, "model.layers.30.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.0625]]]}, "model.layers.30.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[22.5]]]}, "model.layers.30.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.125]]]}, "model.layers.30.self_attn.attn.impl.k_cache": {"inputs": [[[16.875]]]}, "model.layers.30.self_attn.attn.impl.v_cache": {"inputs": [[[1.0625]]]}, "model.layers.30.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.0]], [[15.875]], [[1.0]]], "outputs": [[[0.50390625]], [[1.0]]]}, "model.layers.30.mlp.gate_up_proj": {"inputs": [[[6.5]]], "params": {"weight": [[0.49609375]]}}, "model.layers.30.mlp.down_proj": {"inputs": [[[2.984375]]], "outputs": [[[0.7890625]], [[3.375]]], "params": {"weight": [[1.4453125]]}}, "model.layers.31.self_attn.qkv_proj": {"inputs": [[[12.5]]], "params": {"weight": [[0.298828125]]}}, "model.layers.31.self_attn.o_proj": {"inputs": [[[0.734375]]], "outputs": [[[1.4921875]], [[2.265625]]], "params": {"weight": [[0.361328125]]}}, "model.layers.31.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.8828125]], [[19.125]]]}, "model.layers.31.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.21875]]]}, "model.layers.31.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[52.75]]]}, "model.layers.31.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[19.75]]]}, "model.layers.31.self_attn.attn.impl.k_cache": {"inputs": [[[19.125]]]}, "model.layers.31.self_attn.attn.impl.v_cache": {"inputs": [[[2.21875]]]}, "model.layers.31.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.5625]], [[18.75]], [[2.21875]]], "outputs": [[[0.72265625]], [[1.0]]]}, "model.layers.31.mlp.gate_up_proj": {"inputs": [[[5.09375]]], "params": {"weight": [[0.8046875]]}}, "model.layers.31.mlp.down_proj": {"inputs": [[[1.9609375]]], "outputs": [[[0.60546875]], [[1.5390625]]], "params": {"weight": [[0.74609375]]}}, "model.layers.32.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.357421875]]}}, "model.layers.32.self_attn.o_proj": {"inputs": [[[0.5625]]], "outputs": [[[0.470703125]], [[1.5]]], "params": {"weight": [[0.6484375]]}}, "model.layers.32.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.74609375]], [[13.6875]]]}, "model.layers.32.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.40625]]]}, "model.layers.32.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[34.5]]]}, "model.layers.32.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.5]]]}, "model.layers.32.self_attn.attn.impl.k_cache": {"inputs": [[[13.6875]]]}, "model.layers.32.self_attn.attn.impl.v_cache": {"inputs": [[[4.40625]]]}, "model.layers.32.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.34375]], [[11.75]], [[4.40625]]], "outputs": [[[0.5625]], [[1.0]]]}, "model.layers.32.mlp.gate_up_proj": {"inputs": [[[6.65625]]], "params": {"weight": [[0.3515625]]}}, "model.layers.32.mlp.down_proj": {"inputs": [[[1.7265625]]], "outputs": [[[0.703125]], [[2.703125]]], "params": {"weight": [[0.53515625]]}}, "model.layers.33.self_attn.qkv_proj": {"inputs": [[[14.25]]], "params": {"weight": [[0.27734375]]}}, "model.layers.33.self_attn.o_proj": {"inputs": [[[0.79296875]]], "outputs": [[[1.4921875]], [[2.65625]]], "params": {"weight": [[0.24609375]]}}, "model.layers.33.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.76171875]], [[18.0]]]}, "model.layers.33.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.828125]]]}, "model.layers.33.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[34.25]]]}, "model.layers.33.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.875]]]}, "model.layers.33.self_attn.attn.impl.k_cache": {"inputs": [[[18.0]]]}, "model.layers.33.self_attn.attn.impl.v_cache": {"inputs": [[[2.828125]]]}, "model.layers.33.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.53125]], [[16.5]], [[2.828125]]], "outputs": [[[0.5390625]], [[1.0]]]}, "model.layers.33.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[9.25]]}}, "model.layers.33.mlp.down_proj": {"inputs": [[[1.9453125]]], "outputs": [[[0.890625]], [[1.7578125]]], "params": {"weight": [[0.734375]]}}, "model.layers.34.self_attn.qkv_proj": {"inputs": [[[13.25]]], "params": {"weight": [[0.318359375]]}}, "model.layers.34.self_attn.o_proj": {"inputs": [[[0.765625]]], "outputs": [[[13.5625]], [[14.9375]]], "params": {"weight": [[0.203125]]}}, "model.layers.34.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.90625]], [[16.125]]]}, "model.layers.34.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.734375]]]}, "model.layers.34.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[35.75]]]}, "model.layers.34.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[17.625]]]}, "model.layers.34.self_attn.attn.impl.k_cache": {"inputs": [[[16.125]]]}, "model.layers.34.self_attn.attn.impl.v_cache": {"inputs": [[[2.734375]]]}, "model.layers.34.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.375]], [[15.6875]], [[2.734375]]], "outputs": [[[0.66015625]], [[1.0]]]}, "model.layers.34.mlp.gate_up_proj": {"inputs": [[[7.0]]], "params": {"weight": [[0.318359375]]}}, "model.layers.34.mlp.down_proj": {"inputs": [[[2.5]]], "outputs": [[[0.85546875]], [[3.125]]], "params": {"weight": [[0.55078125]]}}, "model.layers.35.self_attn.qkv_proj": {"inputs": [[[11.8125]]], "params": {"weight": [[0.373046875]]}}, "model.layers.35.self_attn.o_proj": {"inputs": [[[0.90234375]]], "outputs": [[[5.90625]], [[6.71875]]], "params": {"weight": [[0.408203125]]}}, "model.layers.35.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.8125]], [[18.375]]]}, "model.layers.35.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.84375]]]}, "model.layers.35.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[34.0]]]}, "model.layers.35.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.75]]]}, "model.layers.35.self_attn.attn.impl.k_cache": {"inputs": [[[18.375]]]}, "model.layers.35.self_attn.attn.impl.v_cache": {"inputs": [[[2.84375]]]}, "model.layers.35.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.4375]], [[17.0]], [[2.8125]]], "outputs": [[[0.60546875]], [[1.0]]]}, "model.layers.35.mlp.gate_up_proj": {"inputs": [[[6.5]]], "params": {"weight": [[0.26953125]]}}, "model.layers.35.mlp.down_proj": {"inputs": [[[2.640625]]], "outputs": [[[0.8125]], [[4.46875]]], "params": {"weight": [[0.4375]]}}, "model.layers.36.self_attn.qkv_proj": {"inputs": [[[15.4375]]], "params": {"weight": [[0.33984375]]}}, "model.layers.36.self_attn.o_proj": {"inputs": [[[0.5546875]]], "outputs": [[[2.71875]], [[2.75]]], "params": {"weight": [[0.310546875]]}}, "model.layers.36.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6640625]], [[15.8125]]]}, "model.layers.36.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.375]]]}, "model.layers.36.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[34.0]]]}, "model.layers.36.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.125]]]}, "model.layers.36.self_attn.attn.impl.k_cache": {"inputs": [[[15.8125]]]}, "model.layers.36.self_attn.attn.impl.v_cache": {"inputs": [[[2.375]]]}, "model.layers.36.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.3125]], [[14.25]], [[2.375]]], "outputs": [[[0.5546875]], [[1.0]]]}, "model.layers.36.mlp.gate_up_proj": {"inputs": [[[8.125]]], "params": {"weight": [[0.37109375]]}}, "model.layers.36.mlp.down_proj": {"inputs": [[[2.546875]]], "outputs": [[[0.73828125]], [[3.40625]]], "params": {"weight": [[1.46875]]}}, "model.layers.37.self_attn.qkv_proj": {"inputs": [[[17.625]]], "params": {"weight": [[0.33984375]]}}, "model.layers.37.self_attn.o_proj": {"inputs": [[[0.54296875]]], "outputs": [[[2.859375]], [[3.59375]]], "params": {"weight": [[0.18359375]]}}, "model.layers.37.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.76171875]], [[20.0]]]}, "model.layers.37.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.109375]]]}, "model.layers.37.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[47.75]]]}, "model.layers.37.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[18.625]]]}, "model.layers.37.self_attn.attn.impl.k_cache": {"inputs": [[[20.0]]]}, "model.layers.37.self_attn.attn.impl.v_cache": {"inputs": [[[1.109375]]]}, "model.layers.37.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.0625]], [[16.625]], [[0.828125]]], "outputs": [[[0.494140625]], [[1.0]]]}, "model.layers.37.mlp.gate_up_proj": {"inputs": [[[7.4375]]], "params": {"weight": [[0.7265625]]}}, "model.layers.37.mlp.down_proj": {"inputs": [[[2.671875]]], "outputs": [[[0.78515625]], [[2.21875]]], "params": {"weight": [[1.40625]]}}, "model.layers.38.self_attn.qkv_proj": {"inputs": [[[16.125]]], "params": {"weight": [[0.28125]]}}, "model.layers.38.self_attn.o_proj": {"inputs": [[[1.09375]]], "outputs": [[[0.703125]], [[1.109375]]], "params": {"weight": [[0.59375]]}}, "model.layers.38.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6640625]], [[15.125]]]}, "model.layers.38.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.625]]]}, "model.layers.38.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.78125]]]}, "model.layers.38.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.90625]]]}, "model.layers.38.self_attn.attn.impl.k_cache": {"inputs": [[[15.125]]]}, "model.layers.38.self_attn.attn.impl.v_cache": {"inputs": [[[4.625]]]}, "model.layers.38.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.0]], [[14.5]], [[4.625]]], "outputs": [[[1.09375]], [[1.0]]]}, "model.layers.38.mlp.gate_up_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.58984375]]}}, "model.layers.38.mlp.down_proj": {"inputs": [[[2.5]]], "outputs": [[[2.140625]], [[3.078125]]], "params": {"weight": [[0.67578125]]}}, "model.layers.39.self_attn.qkv_proj": {"inputs": [[[17.25]]], "params": {"weight": [[0.5]]}}, "model.layers.39.self_attn.o_proj": {"inputs": [[[0.81640625]]], "outputs": [[[3.515625]], [[3.625]]], "params": {"weight": [[0.10791015625]]}}, "model.layers.39.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.82421875]], [[17.5]]]}, "model.layers.39.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.1953125]]]}, "model.layers.39.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[49.75]]]}, "model.layers.39.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[18.25]]]}, "model.layers.39.self_attn.attn.impl.k_cache": {"inputs": [[[17.5]]]}, "model.layers.39.self_attn.attn.impl.v_cache": {"inputs": [[[1.1953125]]]}, "model.layers.39.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.4375]], [[16.5]], [[1.1953125]]], "outputs": [[[0.5625]], [[1.0]]]}, "model.layers.39.mlp.gate_up_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.318359375]]}}, "model.layers.39.mlp.down_proj": {"inputs": [[[2.46875]]], "outputs": [[[1.6328125]], [[3.203125]]], "params": {"weight": [[0.390625]]}}, "model.layers.40.self_attn.qkv_proj": {"inputs": [[[13.0]]], "params": {"weight": [[0.31640625]]}}, "model.layers.40.self_attn.o_proj": {"inputs": [[[0.9296875]]], "outputs": [[[0.9609375]], [[2.0625]]], "params": {"weight": [[0.259765625]]}}, "model.layers.40.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.73046875]], [[12.5]]]}, "model.layers.40.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.6953125]]]}, "model.layers.40.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[18.5]]]}, "model.layers.40.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[12.5]]]}, "model.layers.40.self_attn.attn.impl.k_cache": {"inputs": [[[12.5]]]}, "model.layers.40.self_attn.attn.impl.v_cache": {"inputs": [[[1.6953125]]]}, "model.layers.40.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.5625]], [[11.625]], [[1.328125]]], "outputs": [[[0.7421875]], [[1.0]]]}, "model.layers.40.mlp.gate_up_proj": {"inputs": [[[7.1875]]], "params": {"weight": [[0.392578125]]}}, "model.layers.40.mlp.down_proj": {"inputs": [[[3.21875]]], "outputs": [[[0.68359375]], [[1.34375]]], "params": {"weight": [[0.44921875]]}}, "model.layers.41.self_attn.qkv_proj": {"inputs": [[[16.25]]], "params": {"weight": [[0.375]]}}, "model.layers.41.self_attn.o_proj": {"inputs": [[[0.486328125]]], "outputs": [[[0.2255859375]], [[1.078125]]], "params": {"weight": [[0.30078125]]}}, "model.layers.41.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.73046875]], [[15.3125]]]}, "model.layers.41.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.96484375]]]}, "model.layers.41.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[21.0]]]}, "model.layers.41.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.9375]]]}, "model.layers.41.self_attn.attn.impl.k_cache": {"inputs": [[[15.3125]]]}, "model.layers.41.self_attn.attn.impl.v_cache": {"inputs": [[[0.96484375]]]}, "model.layers.41.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.53125]], [[14.375]], [[1.1015625]]], "outputs": [[[0.390625]], [[1.0]]]}, "model.layers.41.mlp.gate_up_proj": {"inputs": [[[7.28125]]], "params": {"weight": [[0.294921875]]}}, "model.layers.41.mlp.down_proj": {"inputs": [[[3.203125]]], "outputs": [[[0.9921875]], [[1.3984375]]], "params": {"weight": [[0.2451171875]]}}, "model.layers.42.self_attn.qkv_proj": {"inputs": [[[14.8125]]], "params": {"weight": [[0.494140625]]}}, "model.layers.42.self_attn.o_proj": {"inputs": [[[0.34765625]]], "outputs": [[[0.1474609375]], [[0.5]]], "params": {"weight": [[0.328125]]}}, "model.layers.42.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.53125]], [[11.9375]]]}, "model.layers.42.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.609375]]]}, "model.layers.42.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.84375]]]}, "model.layers.42.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.8125]]]}, "model.layers.42.self_attn.attn.impl.k_cache": {"inputs": [[[11.9375]]]}, "model.layers.42.self_attn.attn.impl.v_cache": {"inputs": [[[3.609375]]]}, "model.layers.42.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.53125]], [[11.9375]], [[3.609375]]], "outputs": [[[0.3359375]], [[1.0]]]}, "model.layers.42.mlp.gate_up_proj": {"inputs": [[[6.75]]], "params": {"weight": [[0.375]]}}, "model.layers.42.mlp.down_proj": {"inputs": [[[3.8125]]], "outputs": [[[0.85546875]], [[2.046875]]], "params": {"weight": [[0.396484375]]}}, "model.layers.43.self_attn.qkv_proj": {"inputs": [[[12.6875]]], "params": {"weight": [[0.384765625]]}}, "model.layers.43.self_attn.o_proj": {"inputs": [[[0.6640625]]], "outputs": [[[0.357421875]], [[0.734375]]], "params": {"weight": [[0.24609375]]}}, "model.layers.43.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.51171875]], [[12.5]]]}, "model.layers.43.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.83203125]]]}, "model.layers.43.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[8.75]]]}, "model.layers.43.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.65625]]]}, "model.layers.43.self_attn.attn.impl.k_cache": {"inputs": [[[12.5]]]}, "model.layers.43.self_attn.attn.impl.v_cache": {"inputs": [[[0.83203125]]]}, "model.layers.43.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.375]], [[11.375]], [[0.83203125]]], "outputs": [[[0.5546875]], [[1.0]]]}, "model.layers.43.mlp.gate_up_proj": {"inputs": [[[6.625]]], "params": {"weight": [[0.40625]]}}, "model.layers.43.mlp.down_proj": {"inputs": [[[9.4375]]], "outputs": [[[0.60546875]], [[0.81640625]]], "params": {"weight": [[0.77734375]]}}, "model.layers.44.self_attn.qkv_proj": {"inputs": [[[13.625]]], "params": {"weight": [[0.3515625]]}}, "model.layers.44.self_attn.o_proj": {"inputs": [[[0.91796875]]], "outputs": [[[0.67578125]], [[0.59375]]], "params": {"weight": [[0.310546875]]}}, "model.layers.44.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.95703125]], [[12.625]]]}, "model.layers.44.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.6015625]]]}, "model.layers.44.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[27.125]]]}, "model.layers.44.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[18.375]]]}, "model.layers.44.self_attn.attn.impl.k_cache": {"inputs": [[[12.625]]]}, "model.layers.44.self_attn.attn.impl.v_cache": {"inputs": [[[1.6015625]]]}, "model.layers.44.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.03125]], [[12.3125]], [[1.546875]]], "outputs": [[[0.6328125]], [[1.0]]]}, "model.layers.44.mlp.gate_up_proj": {"inputs": [[[7.09375]]], "params": {"weight": [[0.5234375]]}}, "model.layers.44.mlp.down_proj": {"inputs": [[[6.65625]]], "outputs": [[[0.486328125]], [[0.984375]]], "params": {"weight": [[0.4375]]}}, "model.layers.45.self_attn.qkv_proj": {"inputs": [[[13.75]]], "params": {"weight": [[0.412109375]]}}, "model.layers.45.self_attn.o_proj": {"inputs": [[[0.5859375]]], "outputs": [[[0.546875]], [[0.9921875]]], "params": {"weight": [[0.205078125]]}}, "model.layers.45.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.57421875]], [[12.125]]]}, "model.layers.45.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.0625]]]}, "model.layers.45.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[14.3125]]]}, "model.layers.45.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.25]]]}, "model.layers.45.self_attn.attn.impl.k_cache": {"inputs": [[[12.125]]]}, "model.layers.45.self_attn.attn.impl.v_cache": {"inputs": [[[1.0625]]]}, "model.layers.45.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.65625]], [[11.8125]], [[0.8984375]]], "outputs": [[[0.46484375]], [[1.0]]]}, "model.layers.45.mlp.gate_up_proj": {"inputs": [[[7.15625]]], "params": {"weight": [[0.6484375]]}}, "model.layers.45.mlp.down_proj": {"inputs": [[[8.625]]], "outputs": [[[1.515625]], [[2.4375]]], "params": {"weight": [[0.38671875]]}}, "model.layers.46.self_attn.qkv_proj": {"inputs": [[[12.9375]]], "params": {"weight": [[0.73046875]]}}, "model.layers.46.self_attn.o_proj": {"inputs": [[[0.384765625]]], "outputs": [[[0.14453125]], [[0.2099609375]]], "params": {"weight": [[0.1689453125]]}}, "model.layers.46.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.67578125]], [[15.625]]]}, "model.layers.46.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.609375]]]}, "model.layers.46.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[15.125]]]}, "model.layers.46.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.84375]]]}, "model.layers.46.self_attn.attn.impl.k_cache": {"inputs": [[[15.625]]]}, "model.layers.46.self_attn.attn.impl.v_cache": {"inputs": [[[0.609375]]]}, "model.layers.46.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.0]], [[15.625]], [[0.609375]]], "outputs": [[[0.384765625]], [[1.0]]]}, "model.layers.46.mlp.gate_up_proj": {"inputs": [[[6.4375]]], "params": {"weight": [[0.32421875]]}}, "model.layers.46.mlp.down_proj": {"inputs": [[[4.15625]]], "outputs": [[[0.359375]], [[0.671875]]], "params": {"weight": [[0.251953125]]}}, "model.layers.47.self_attn.qkv_proj": {"inputs": [[[11.9375]]], "params": {"weight": [[0.431640625]]}}, "model.layers.47.self_attn.o_proj": {"inputs": [[[0.6640625]]], "outputs": [[[0.625]], [[0.64453125]]], "params": {"weight": [[0.244140625]]}}, "model.layers.47.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.60546875]], [[12.3125]]]}, "model.layers.47.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.609375]]]}, "model.layers.47.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[29.875]]]}, "model.layers.47.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.125]]]}, "model.layers.47.self_attn.attn.impl.k_cache": {"inputs": [[[12.3125]]]}, "model.layers.47.self_attn.attn.impl.v_cache": {"inputs": [[[3.609375]]]}, "model.layers.47.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.21875]], [[11.625]], [[3.609375]]], "outputs": [[[0.431640625]], [[1.0]]]}, "model.layers.47.mlp.gate_up_proj": {"inputs": [[[6.5625]]], "params": {"weight": [[0.51953125]]}}, "model.layers.47.mlp.down_proj": {"inputs": [[[6.15625]]], "outputs": [[[0.5234375]], [[1.21875]]], "params": {"weight": [[0.236328125]]}}, "model.layers.48.self_attn.qkv_proj": {"inputs": [[[11.125]]], "params": {"weight": [[0.427734375]]}}, "model.layers.48.self_attn.o_proj": {"inputs": [[[0.72265625]]], "outputs": [[[0.3984375]], [[0.515625]]], "params": {"weight": [[0.1796875]]}}, "model.layers.48.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.54296875]], [[13.625]]]}, "model.layers.48.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.2578125]]]}, "model.layers.48.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[6.5625]]]}, "model.layers.48.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.5625]]]}, "model.layers.48.self_attn.attn.impl.k_cache": {"inputs": [[[13.625]]]}, "model.layers.48.self_attn.attn.impl.v_cache": {"inputs": [[[1.2578125]]]}, "model.layers.48.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.53125]], [[13.0625]], [[1.1953125]]], "outputs": [[[0.72265625]], [[1.0]]]}, "model.layers.48.mlp.gate_up_proj": {"inputs": [[[6.84375]]], "params": {"weight": [[0.5703125]]}}, "model.layers.48.mlp.down_proj": {"inputs": [[[4.8125]]], "outputs": [[[0.447265625]], [[2.171875]]], "params": {"weight": [[0.462890625]]}}, "model.layers.49.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.46484375]]}}, "model.layers.49.self_attn.o_proj": {"inputs": [[[0.443359375]]], "outputs": [[[0.333984375]], [[0.4140625]]], "params": {"weight": [[0.232421875]]}}, "model.layers.49.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.63671875]], [[11.5625]]]}, "model.layers.49.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.77734375]]]}, "model.layers.49.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[8.6875]]]}, "model.layers.49.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.75]]]}, "model.layers.49.self_attn.attn.impl.k_cache": {"inputs": [[[11.5625]]]}, "model.layers.49.self_attn.attn.impl.v_cache": {"inputs": [[[0.77734375]]]}, "model.layers.49.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.625]], [[11.375]], [[0.76953125]]], "outputs": [[[0.361328125]], [[1.0]]]}, "model.layers.49.mlp.gate_up_proj": {"inputs": [[[6.6875]]], "params": {"weight": [[0.45703125]]}}, "model.layers.49.mlp.down_proj": {"inputs": [[[3.9375]]], "outputs": [[[1.15625]], [[3.046875]]], "params": {"weight": [[0.24609375]]}}, "model.layers.50.self_attn.qkv_proj": {"inputs": [[[11.125]]], "params": {"weight": [[0.486328125]]}}, "model.layers.50.self_attn.o_proj": {"inputs": [[[0.7734375]]], "outputs": [[[0.2080078125]], [[0.4453125]]], "params": {"weight": [[0.1826171875]]}}, "model.layers.50.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.25]], [[11.75]]]}, "model.layers.50.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.546875]]]}, "model.layers.50.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[26.75]]]}, "model.layers.50.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.6875]]]}, "model.layers.50.self_attn.attn.impl.k_cache": {"inputs": [[[11.75]]]}, "model.layers.50.self_attn.attn.impl.v_cache": {"inputs": [[[1.546875]]]}, "model.layers.50.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[12.4375]], [[11.0625]], [[1.484375]]], "outputs": [[[0.59375]], [[1.0]]]}, "model.layers.50.mlp.gate_up_proj": {"inputs": [[[6.53125]]], "params": {"weight": [[0.45703125]]}}, "model.layers.50.mlp.down_proj": {"inputs": [[[9.6875]]], "outputs": [[[0.92578125]], [[0.8125]]], "params": {"weight": [[0.44140625]]}}, "model.layers.51.self_attn.qkv_proj": {"inputs": [[[10.875]]], "params": {"weight": [[0.77734375]]}}, "model.layers.51.self_attn.o_proj": {"inputs": [[[0.51171875]]], "outputs": [[[0.396484375]], [[0.578125]]], "params": {"weight": [[0.3359375]]}}, "model.layers.51.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.54296875]], [[12.5]]]}, "model.layers.51.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.8125]]]}, "model.layers.51.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[10.6875]]]}, "model.layers.51.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.625]]]}, "model.layers.51.self_attn.attn.impl.k_cache": {"inputs": [[[12.5]]]}, "model.layers.51.self_attn.attn.impl.v_cache": {"inputs": [[[4.8125]]]}, "model.layers.51.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.03125]], [[12.5]], [[4.8125]]], "outputs": [[[0.51171875]], [[1.0]]]}, "model.layers.51.mlp.gate_up_proj": {"inputs": [[[6.46875]]], "params": {"weight": [[0.40625]]}}, "model.layers.51.mlp.down_proj": {"inputs": [[[8.9375]]], "outputs": [[[0.44921875]], [[2.796875]]], "params": {"weight": [[0.4296875]]}}, "model.layers.52.self_attn.qkv_proj": {"inputs": [[[12.375]]], "params": {"weight": [[0.33203125]]}}, "model.layers.52.self_attn.o_proj": {"inputs": [[[0.63671875]]], "outputs": [[[0.671875]], [[0.9375]]], "params": {"weight": [[0.306640625]]}}, "model.layers.52.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.73828125]], [[13.8125]]]}, "model.layers.52.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.3046875]]]}, "model.layers.52.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[22.5]]]}, "model.layers.52.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.5]]]}, "model.layers.52.self_attn.attn.impl.k_cache": {"inputs": [[[13.8125]]]}, "model.layers.52.self_attn.attn.impl.v_cache": {"inputs": [[[1.3046875]]]}, "model.layers.52.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.21875]], [[12.625]], [[1.265625]]], "outputs": [[[0.6328125]], [[1.0]]]}, "model.layers.52.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[0.412109375]]}}, "model.layers.52.mlp.down_proj": {"inputs": [[[3.296875]]], "outputs": [[[0.85546875]], [[1.9453125]]], "params": {"weight": [[0.271484375]]}}, "model.layers.53.self_attn.qkv_proj": {"inputs": [[[11.25]]], "params": {"weight": [[0.419921875]]}}, "model.layers.53.self_attn.o_proj": {"inputs": [[[0.67578125]]], "outputs": [[[0.279296875]], [[0.63671875]]], "params": {"weight": [[0.212890625]]}}, "model.layers.53.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.63671875]], [[13.75]]]}, "model.layers.53.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.2578125]]]}, "model.layers.53.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[10.875]]]}, "model.layers.53.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.59375]]]}, "model.layers.53.self_attn.attn.impl.k_cache": {"inputs": [[[13.75]]]}, "model.layers.53.self_attn.attn.impl.v_cache": {"inputs": [[[1.2578125]]]}, "model.layers.53.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.34375]], [[12.875]], [[1.2578125]]], "outputs": [[[0.494140625]], [[1.0]]]}, "model.layers.53.mlp.gate_up_proj": {"inputs": [[[6.6875]]], "params": {"weight": [[0.416015625]]}}, "model.layers.53.mlp.down_proj": {"inputs": [[[5.21875]]], "outputs": [[[0.59765625]], [[1.921875]]], "params": {"weight": [[1.1796875]]}}, "model.layers.54.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.59375]]}}, "model.layers.54.self_attn.o_proj": {"inputs": [[[0.43359375]]], "outputs": [[[0.66796875]], [[1.078125]]], "params": {"weight": [[0.19140625]]}}, "model.layers.54.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.609375]], [[12.875]]]}, "model.layers.54.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.2734375]]]}, "model.layers.54.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.125]]]}, "model.layers.54.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.4375]]]}, "model.layers.54.self_attn.attn.impl.k_cache": {"inputs": [[[12.875]]]}, "model.layers.54.self_attn.attn.impl.v_cache": {"inputs": [[[1.2734375]]]}, "model.layers.54.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.75]], [[12.875]], [[1.2734375]]], "outputs": [[[0.345703125]], [[1.0]]]}, "model.layers.54.mlp.gate_up_proj": {"inputs": [[[6.5625]]], "params": {"weight": [[0.82421875]]}}, "model.layers.54.mlp.down_proj": {"inputs": [[[8.0625]]], "outputs": [[[0.6171875]], [[1.2734375]]], "params": {"weight": [[1.2890625]]}}, "model.layers.55.self_attn.qkv_proj": {"inputs": [[[10.3125]]], "params": {"weight": [[0.58203125]]}}, "model.layers.55.self_attn.o_proj": {"inputs": [[[0.74609375]]], "outputs": [[[0.9765625]], [[1.28125]]], "params": {"weight": [[0.2001953125]]}}, "model.layers.55.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.39453125]], [[16.0]]]}, "model.layers.55.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.3828125]]]}, "model.layers.55.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[11.625]]]}, "model.layers.55.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.125]]]}, "model.layers.55.self_attn.attn.impl.k_cache": {"inputs": [[[16.0]]]}, "model.layers.55.self_attn.attn.impl.v_cache": {"inputs": [[[1.3828125]]]}, "model.layers.55.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.5]], [[12.3125]], [[1.3828125]]], "outputs": [[[0.74609375]], [[1.0]]]}, "model.layers.55.mlp.gate_up_proj": {"inputs": [[[6.6875]]], "params": {"weight": [[0.66015625]]}}, "model.layers.55.mlp.down_proj": {"inputs": [[[5.375]]], "outputs": [[[1.59375]], [[3.078125]]], "params": {"weight": [[0.408203125]]}}, "model.layers.56.self_attn.qkv_proj": {"inputs": [[[13.6875]]], "params": {"weight": [[0.376953125]]}}, "model.layers.56.self_attn.o_proj": {"inputs": [[[1.4921875]]], "outputs": [[[0.72265625]], [[1.0546875]]], "params": {"weight": [[0.09912109375]]}}, "model.layers.56.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.296875]], [[13.75]]]}, "model.layers.56.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.28125]]]}, "model.layers.56.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[24.25]]]}, "model.layers.56.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[12.8125]]]}, "model.layers.56.self_attn.attn.impl.k_cache": {"inputs": [[[13.75]]]}, "model.layers.56.self_attn.attn.impl.v_cache": {"inputs": [[[2.28125]]]}, "model.layers.56.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[13.1875]], [[13.0]], [[1.6328125]]], "outputs": [[[0.5390625]], [[1.0]]]}, "model.layers.56.mlp.gate_up_proj": {"inputs": [[[6.90625]]], "params": {"weight": [[0.41015625]]}}, "model.layers.56.mlp.down_proj": {"inputs": [[[6.875]]], "outputs": [[[2.109375]], [[2.296875]]], "params": {"weight": [[0.275390625]]}}, "model.layers.57.self_attn.qkv_proj": {"inputs": [[[12.1875]]], "params": {"weight": [[0.466796875]]}}, "model.layers.57.self_attn.o_proj": {"inputs": [[[0.71875]]], "outputs": [[[0.306640625]], [[0.7578125]]], "params": {"weight": [[0.22265625]]}}, "model.layers.57.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5546875]], [[14.5625]]]}, "model.layers.57.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.2109375]]]}, "model.layers.57.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[8.75]]]}, "model.layers.57.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.28125]]]}, "model.layers.57.self_attn.attn.impl.k_cache": {"inputs": [[[14.5625]]]}, "model.layers.57.self_attn.attn.impl.v_cache": {"inputs": [[[1.2109375]]]}, "model.layers.57.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.21875]], [[14.5625]], [[1.0390625]]], "outputs": [[[0.53515625]], [[1.0]]]}, "model.layers.57.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[0.4140625]]}}, "model.layers.57.mlp.down_proj": {"inputs": [[[5.59375]]], "outputs": [[[0.66015625]], [[1.9609375]]], "params": {"weight": [[0.578125]]}}, "model.layers.58.self_attn.qkv_proj": {"inputs": [[[12.1875]]], "params": {"weight": [[0.484375]]}}, "model.layers.58.self_attn.o_proj": {"inputs": [[[0.70703125]]], "outputs": [[[0.298828125]], [[0.66796875]]], "params": {"weight": [[0.1923828125]]}}, "model.layers.58.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.59375]], [[17.375]]]}, "model.layers.58.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.77734375]]]}, "model.layers.58.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[11.875]]]}, "model.layers.58.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.3125]]]}, "model.layers.58.self_attn.attn.impl.k_cache": {"inputs": [[[17.375]]]}, "model.layers.58.self_attn.attn.impl.v_cache": {"inputs": [[[0.77734375]]]}, "model.layers.58.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.15625]], [[14.0]], [[0.7578125]]], "outputs": [[[0.48828125]], [[1.0]]]}, "model.layers.58.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[0.271484375]]}}, "model.layers.58.mlp.down_proj": {"inputs": [[[6.0]]], "outputs": [[[0.423828125]], [[0.734375]]], "params": {"weight": [[0.296875]]}}, "model.layers.59.self_attn.qkv_proj": {"inputs": [[[10.8125]]], "params": {"weight": [[0.6015625]]}}, "model.layers.59.self_attn.o_proj": {"inputs": [[[0.57421875]]], "outputs": [[[0.251953125]], [[1.0625]]], "params": {"weight": [[0.228515625]]}}, "model.layers.59.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5078125]], [[11.4375]]]}, "model.layers.59.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.1328125]]]}, "model.layers.59.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[22.875]]]}, "model.layers.59.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.875]]]}, "model.layers.59.self_attn.attn.impl.k_cache": {"inputs": [[[11.4375]]]}, "model.layers.59.self_attn.attn.impl.v_cache": {"inputs": [[[1.1328125]]]}, "model.layers.59.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.0625]], [[11.4375]], [[1.1328125]]], "outputs": [[[0.5078125]], [[1.0]]]}, "model.layers.59.mlp.gate_up_proj": {"inputs": [[[7.21875]]], "params": {"weight": [[0.3125]]}}, "model.layers.59.mlp.down_proj": {"inputs": [[[7.40625]]], "outputs": [[[0.578125]], [[7.1875]]], "params": {"weight": [[0.490234375]]}}, "model.layers.60.self_attn.qkv_proj": {"inputs": [[[13.3125]]], "params": {"weight": [[0.400390625]]}}, "model.layers.60.self_attn.o_proj": {"inputs": [[[1.1328125]]], "outputs": [[[0.484375]], [[0.58984375]]], "params": {"weight": [[0.20703125]]}}, "model.layers.60.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.50390625]], [[17.75]]]}, "model.layers.60.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.078125]]]}, "model.layers.60.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[8.1875]]]}, "model.layers.60.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.78125]]]}, "model.layers.60.self_attn.attn.impl.k_cache": {"inputs": [[[17.75]]]}, "model.layers.60.self_attn.attn.impl.v_cache": {"inputs": [[[2.078125]]]}, "model.layers.60.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.46875]], [[11.9375]], [[1.2578125]]], "outputs": [[[0.734375]], [[1.0]]]}, "model.layers.60.mlp.gate_up_proj": {"inputs": [[[7.34375]]], "params": {"weight": [[0.373046875]]}}, "model.layers.60.mlp.down_proj": {"inputs": [[[13.375]]], "outputs": [[[2.296875]], [[3.671875]]], "params": {"weight": [[1.5]]}}, "model.layers.61.self_attn.qkv_proj": {"inputs": [[[11.5]]], "params": {"weight": [[0.58984375]]}}, "model.layers.61.self_attn.o_proj": {"inputs": [[[0.4609375]]], "outputs": [[[0.26953125]], [[0.62109375]]], "params": {"weight": [[0.1982421875]]}}, "model.layers.61.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5859375]], [[15.6875]]]}, "model.layers.61.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.4140625]]]}, "model.layers.61.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.375]]]}, "model.layers.61.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.0625]]]}, "model.layers.61.self_attn.attn.impl.k_cache": {"inputs": [[[15.6875]]]}, "model.layers.61.self_attn.attn.impl.v_cache": {"inputs": [[[1.4140625]]]}, "model.layers.61.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.3125]], [[15.6875]], [[1.3359375]]], "outputs": [[[0.4609375]], [[1.0]]]}, "model.layers.61.mlp.gate_up_proj": {"inputs": [[[7.46875]]], "params": {"weight": [[0.322265625]]}}, "model.layers.61.mlp.down_proj": {"inputs": [[[8.75]]], "outputs": [[[0.7421875]], [[2.421875]]], "params": {"weight": [[0.412109375]]}}, "model.layers.62.self_attn.qkv_proj": {"inputs": [[[11.5625]]], "params": {"weight": [[0.9453125]]}}, "model.layers.62.self_attn.o_proj": {"inputs": [[[0.478515625]]], "outputs": [[[0.75]], [[0.76171875]]], "params": {"weight": [[0.12158203125]]}}, "model.layers.62.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.71875]], [[25.125]]]}, "model.layers.62.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.64453125]]]}, "model.layers.62.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[13.5]]]}, "model.layers.62.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.5]]]}, "model.layers.62.self_attn.attn.impl.k_cache": {"inputs": [[[25.125]]]}, "model.layers.62.self_attn.attn.impl.v_cache": {"inputs": [[[0.64453125]]]}, "model.layers.62.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.6875]], [[25.125]], [[0.77734375]]], "outputs": [[[0.443359375]], [[1.0]]]}, "model.layers.62.mlp.gate_up_proj": {"inputs": [[[7.625]]], "params": {"weight": [[0.322265625]]}}, "model.layers.62.mlp.down_proj": {"inputs": [[[8.5]]], "outputs": [[[1.0078125]], [[1.3359375]]], "params": {"weight": [[0.765625]]}}, "model.layers.63.self_attn.qkv_proj": {"inputs": [[[10.1875]]], "params": {"weight": [[0.65625]]}}, "model.layers.63.self_attn.o_proj": {"inputs": [[[0.482421875]]], "outputs": [[[0.259765625]], [[1.1875]]], "params": {"weight": [[0.310546875]]}}, "model.layers.63.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.515625]], [[13.5625]]]}, "model.layers.63.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.390625]]]}, "model.layers.63.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.875]]]}, "model.layers.63.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.6875]]]}, "model.layers.63.self_attn.attn.impl.k_cache": {"inputs": [[[13.5625]]]}, "model.layers.63.self_attn.attn.impl.v_cache": {"inputs": [[[1.390625]]]}, "model.layers.63.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.78125]], [[12.625]], [[1.390625]]], "outputs": [[[0.353515625]], [[1.0]]]}, "model.layers.63.mlp.gate_up_proj": {"inputs": [[[7.9375]]], "params": {"weight": [[0.4375]]}}, "model.layers.63.mlp.down_proj": {"inputs": [[[6.875]]], "outputs": [[[0.73828125]], [[1.640625]]], "params": {"weight": [[0.1865234375]]}}, "model.layers.64.self_attn.qkv_proj": {"inputs": [[[13.0]]], "params": {"weight": [[0.427734375]]}}, "model.layers.64.self_attn.o_proj": {"inputs": [[[0.62890625]]], "outputs": [[[0.341796875]], [[1.5]]], "params": {"weight": [[0.34375]]}}, "model.layers.64.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.57421875]], [[13.0625]]]}, "model.layers.64.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.0625]]]}, "model.layers.64.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[11.0625]]]}, "model.layers.64.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.28125]]]}, "model.layers.64.self_attn.attn.impl.k_cache": {"inputs": [[[13.0625]]]}, "model.layers.64.self_attn.attn.impl.v_cache": {"inputs": [[[1.0625]]]}, "model.layers.64.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.875]], [[13.0625]], [[0.83984375]]], "outputs": [[[0.62890625]], [[1.0]]]}, "model.layers.64.mlp.gate_up_proj": {"inputs": [[[8.3125]]], "params": {"weight": [[0.484375]]}}, "model.layers.64.mlp.down_proj": {"inputs": [[[5.78125]]], "outputs": [[[0.578125]], [[2.578125]]], "params": {"weight": [[0.482421875]]}}, "model.layers.65.self_attn.qkv_proj": {"inputs": [[[14.6875]]], "params": {"weight": [[0.45703125]]}}, "model.layers.65.self_attn.o_proj": {"inputs": [[[1.6484375]]], "outputs": [[[1.75]], [[1.6875]]], "params": {"weight": [[0.10986328125]]}}, "model.layers.65.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5234375]], [[12.75]]]}, "model.layers.65.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.84375]]]}, "model.layers.65.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[9.0625]]]}, "model.layers.65.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.71875]]]}, "model.layers.65.self_attn.attn.impl.k_cache": {"inputs": [[[12.75]]]}, "model.layers.65.self_attn.attn.impl.v_cache": {"inputs": [[[1.84375]]]}, "model.layers.65.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.59375]], [[11.625]], [[1.0]]], "outputs": [[[0.59375]], [[1.0]]]}, "model.layers.65.mlp.gate_up_proj": {"inputs": [[[8.125]]], "params": {"weight": [[0.419921875]]}}, "model.layers.65.mlp.down_proj": {"inputs": [[[9.9375]]], "outputs": [[[0.76171875]], [[2.015625]]], "params": {"weight": [[0.5234375]]}}, "model.layers.66.self_attn.qkv_proj": {"inputs": [[[11.375]]], "params": {"weight": [[0.49609375]]}}, "model.layers.66.self_attn.o_proj": {"inputs": [[[0.54296875]]], "outputs": [[[1.109375]], [[1.3359375]]], "params": {"weight": [[0.10205078125]]}}, "model.layers.66.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5234375]], [[13.4375]]]}, "model.layers.66.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.9375]]]}, "model.layers.66.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.0]]]}, "model.layers.66.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.09375]]]}, "model.layers.66.self_attn.attn.impl.k_cache": {"inputs": [[[13.4375]]]}, "model.layers.66.self_attn.attn.impl.v_cache": {"inputs": [[[0.9375]]]}, "model.layers.66.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.8125]], [[11.5625]], [[0.9375]]], "outputs": [[[0.5234375]], [[1.0]]]}, "model.layers.66.mlp.gate_up_proj": {"inputs": [[[7.78125]]], "params": {"weight": [[0.359375]]}}, "model.layers.66.mlp.down_proj": {"inputs": [[[8.375]]], "outputs": [[[2.5625]], [[3.5]]], "params": {"weight": [[1.28125]]}}, "model.layers.67.self_attn.qkv_proj": {"inputs": [[[12.6875]]], "params": {"weight": [[0.490234375]]}}, "model.layers.67.self_attn.o_proj": {"inputs": [[[0.609375]]], "outputs": [[[0.62890625]], [[1.453125]]], "params": {"weight": [[0.2177734375]]}}, "model.layers.67.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.64453125]], [[13.0]]]}, "model.layers.67.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.78125]]]}, "model.layers.67.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.875]]]}, "model.layers.67.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.75]]]}, "model.layers.67.self_attn.attn.impl.k_cache": {"inputs": [[[13.0]]]}, "model.layers.67.self_attn.attn.impl.v_cache": {"inputs": [[[3.78125]]]}, "model.layers.67.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.53125]], [[12.0]], [[3.78125]]], "outputs": [[[0.56640625]], [[1.0]]]}, "model.layers.67.mlp.gate_up_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.41015625]]}}, "model.layers.67.mlp.down_proj": {"inputs": [[[4.84375]]], "outputs": [[[0.68359375]], [[1.5546875]]], "params": {"weight": [[0.3125]]}}, "model.layers.68.self_attn.qkv_proj": {"inputs": [[[13.75]]], "params": {"weight": [[0.51953125]]}}, "model.layers.68.self_attn.o_proj": {"inputs": [[[1.84375]]], "outputs": [[[0.294921875]], [[2.09375]]], "params": {"weight": [[0.130859375]]}}, "model.layers.68.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.75]], [[14.0625]]]}, "model.layers.68.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.109375]]]}, "model.layers.68.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[15.6875]]]}, "model.layers.68.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.0625]]]}, "model.layers.68.self_attn.attn.impl.k_cache": {"inputs": [[[14.0625]]]}, "model.layers.68.self_attn.attn.impl.v_cache": {"inputs": [[[2.109375]]]}, "model.layers.68.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.65625]], [[13.25]], [[1.28125]]], "outputs": [[[0.8515625]], [[1.0]]]}, "model.layers.68.mlp.gate_up_proj": {"inputs": [[[9.0625]]], "params": {"weight": [[0.470703125]]}}, "model.layers.68.mlp.down_proj": {"inputs": [[[6.125]]], "outputs": [[[1.6171875]], [[3.953125]]], "params": {"weight": [[0.234375]]}}, "model.layers.69.self_attn.qkv_proj": {"inputs": [[[12.0]]], "params": {"weight": [[0.396484375]]}}, "model.layers.69.self_attn.o_proj": {"inputs": [[[1.4609375]]], "outputs": [[[0.58203125]], [[2.109375]]], "params": {"weight": [[0.291015625]]}}, "model.layers.69.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.66796875]], [[17.5]]]}, "model.layers.69.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.109375]]]}, "model.layers.69.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[14.625]]]}, "model.layers.69.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[12.1875]]]}, "model.layers.69.self_attn.attn.impl.k_cache": {"inputs": [[[17.5]]]}, "model.layers.69.self_attn.attn.impl.v_cache": {"inputs": [[[2.109375]]]}, "model.layers.69.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.875]], [[15.0]], [[2.109375]]], "outputs": [[[1.4296875]], [[1.0]]]}, "model.layers.69.mlp.gate_up_proj": {"inputs": [[[8.625]]], "params": {"weight": [[0.419921875]]}}, "model.layers.69.mlp.down_proj": {"inputs": [[[12.5625]]], "outputs": [[[0.875]], [[1.1015625]]], "params": {"weight": [[0.3203125]]}}, "model.layers.70.self_attn.qkv_proj": {"inputs": [[[12.625]]], "params": {"weight": [[0.298828125]]}}, "model.layers.70.self_attn.o_proj": {"inputs": [[[0.9921875]]], "outputs": [[[0.484375]], [[2.5]]], "params": {"weight": [[0.7578125]]}}, "model.layers.70.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.68359375]], [[15.5]]]}, "model.layers.70.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.71875]]]}, "model.layers.70.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[13.4375]]]}, "model.layers.70.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.34375]]]}, "model.layers.70.self_attn.attn.impl.k_cache": {"inputs": [[[15.5]]]}, "model.layers.70.self_attn.attn.impl.v_cache": {"inputs": [[[1.71875]]]}, "model.layers.70.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.15625]], [[14.0]], [[1.703125]]], "outputs": [[[0.859375]], [[1.0]]]}, "model.layers.70.mlp.gate_up_proj": {"inputs": [[[8.875]]], "params": {"weight": [[0.54296875]]}}, "model.layers.70.mlp.down_proj": {"inputs": [[[12.6875]]], "outputs": [[[1.625]], [[4.875]]], "params": {"weight": [[1.0546875]]}}, "model.layers.71.self_attn.qkv_proj": {"inputs": [[[13.0625]]], "params": {"weight": [[0.302734375]]}}, "model.layers.71.self_attn.o_proj": {"inputs": [[[2.484375]]], "outputs": [[[1.125]], [[5.34375]]], "params": {"weight": [[0.54296875]]}}, "model.layers.71.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.0859375]], [[15.0]]]}, "model.layers.71.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.109375]]]}, "model.layers.71.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[11.0]]]}, "model.layers.71.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.78125]]]}, "model.layers.71.self_attn.attn.impl.k_cache": {"inputs": [[[15.0]]]}, "model.layers.71.self_attn.attn.impl.v_cache": {"inputs": [[[3.109375]]]}, "model.layers.71.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[11.75]], [[14.0]], [[2.796875]]], "outputs": [[[1.5625]], [[1.0]]]}, "model.layers.71.mlp.gate_up_proj": {"inputs": [[[9.4375]]], "params": {"weight": [[0.375]]}}, "model.layers.71.mlp.down_proj": {"inputs": [[[6.46875]]], "outputs": [[[1.703125]], [[3.953125]]], "params": {"weight": [[0.69921875]]}}, "model.layers.72.self_attn.qkv_proj": {"inputs": [[[11.6875]]], "params": {"weight": [[0.42578125]]}}, "model.layers.72.self_attn.o_proj": {"inputs": [[[1.984375]]], "outputs": [[[0.81640625]], [[3.484375]]], "params": {"weight": [[0.265625]]}}, "model.layers.72.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.921875]], [[20.875]]]}, "model.layers.72.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.390625]]]}, "model.layers.72.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.5]]]}, "model.layers.72.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.0625]]]}, "model.layers.72.self_attn.attn.impl.k_cache": {"inputs": [[[20.875]]]}, "model.layers.72.self_attn.attn.impl.v_cache": {"inputs": [[[3.390625]]]}, "model.layers.72.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.6875]], [[20.375]], [[3.21875]]], "outputs": [[[1.953125]], [[1.0]]]}, "model.layers.72.mlp.gate_up_proj": {"inputs": [[[9.75]]], "params": {"weight": [[0.427734375]]}}, "model.layers.72.mlp.down_proj": {"inputs": [[[7.9375]]], "outputs": [[[1.40625]], [[2.53125]]], "params": {"weight": [[0.6328125]]}}, "model.layers.73.self_attn.qkv_proj": {"inputs": [[[12.0]]], "params": {"weight": [[0.322265625]]}}, "model.layers.73.self_attn.o_proj": {"inputs": [[[1.640625]]], "outputs": [[[2.40625]], [[3.78125]]], "params": {"weight": [[0.224609375]]}}, "model.layers.73.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.0546875]], [[13.3125]]]}, "model.layers.73.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.703125]]]}, "model.layers.73.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[8.875]]]}, "model.layers.73.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.15625]]]}, "model.layers.73.self_attn.attn.impl.k_cache": {"inputs": [[[13.3125]]]}, "model.layers.73.self_attn.attn.impl.v_cache": {"inputs": [[[2.703125]]]}, "model.layers.73.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[12.625]], [[12.6875]], [[2.734375]]], "outputs": [[[0.76953125]], [[1.0]]]}, "model.layers.73.mlp.gate_up_proj": {"inputs": [[[9.9375]]], "params": {"weight": [[0.66015625]]}}, "model.layers.73.mlp.down_proj": {"inputs": [[[7.125]]], "outputs": [[[2.78125]], [[4.25]]], "params": {"weight": [[0.55078125]]}}, "model.layers.74.self_attn.qkv_proj": {"inputs": [[[11.625]]], "params": {"weight": [[0.2470703125]]}}, "model.layers.74.self_attn.o_proj": {"inputs": [[[1.4609375]]], "outputs": [[[3.15625]], [[5.59375]]], "params": {"weight": [[0.78515625]]}}, "model.layers.74.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.78515625]], [[18.75]]]}, "model.layers.74.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.28125]]]}, "model.layers.74.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[23.375]]]}, "model.layers.74.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.625]]]}, "model.layers.74.self_attn.attn.impl.k_cache": {"inputs": [[[18.75]]]}, "model.layers.74.self_attn.attn.impl.v_cache": {"inputs": [[[2.28125]]]}, "model.layers.74.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.75]], [[18.375]], [[2.28125]]], "outputs": [[[0.84765625]], [[1.0]]]}, "model.layers.74.mlp.gate_up_proj": {"inputs": [[[8.8125]]], "params": {"weight": [[0.390625]]}}, "model.layers.74.mlp.down_proj": {"inputs": [[[11.0625]]], "outputs": [[[2.53125]], [[6.9375]]], "params": {"weight": [[0.703125]]}}, "model.layers.75.self_attn.qkv_proj": {"inputs": [[[11.5625]]], "params": {"weight": [[0.33984375]]}}, "model.layers.75.self_attn.o_proj": {"inputs": [[[1.8984375]]], "outputs": [[[1.25]], [[6.65625]]], "params": {"weight": [[0.193359375]]}}, "model.layers.75.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.86328125]], [[17.875]]]}, "model.layers.75.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.515625]]]}, "model.layers.75.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[30.375]]]}, "model.layers.75.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.625]]]}, "model.layers.75.self_attn.attn.impl.k_cache": {"inputs": [[[17.875]]]}, "model.layers.75.self_attn.attn.impl.v_cache": {"inputs": [[[2.515625]]]}, "model.layers.75.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[11.5]], [[16.375]], [[1.9921875]]], "outputs": [[[1.09375]], [[1.0]]]}, "model.layers.75.mlp.gate_up_proj": {"inputs": [[[8.8125]]], "params": {"weight": [[0.427734375]]}}, "model.layers.75.mlp.down_proj": {"inputs": [[[7.90625]]], "outputs": [[[9.875]], [[12.4375]]], "params": {"weight": [[0.70703125]]}}, "model.layers.76.self_attn.qkv_proj": {"inputs": [[[9.0]]], "params": {"weight": [[0.2421875]]}}, "model.layers.76.self_attn.o_proj": {"inputs": [[[1.40625]]], "outputs": [[[3.03125]], [[5.3125]]], "params": {"weight": [[0.43359375]]}}, "model.layers.76.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.03125]], [[17.125]]]}, "model.layers.76.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.046875]]]}, "model.layers.76.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[19.875]]]}, "model.layers.76.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.0625]]]}, "model.layers.76.self_attn.attn.impl.k_cache": {"inputs": [[[17.125]]]}, "model.layers.76.self_attn.attn.impl.v_cache": {"inputs": [[[3.046875]]]}, "model.layers.76.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.8125]], [[15.5625]], [[3.046875]]], "outputs": [[[1.3984375]], [[1.0]]]}, "model.layers.76.mlp.gate_up_proj": {"inputs": [[[9.375]]], "params": {"weight": [[0.478515625]]}}, "model.layers.76.mlp.down_proj": {"inputs": [[[12.75]]], "outputs": [[[2.078125]], [[4.90625]]], "params": {"weight": [[0.66796875]]}}, "model.layers.77.self_attn.qkv_proj": {"inputs": [[[9.375]]], "params": {"weight": [[0.349609375]]}}, "model.layers.77.self_attn.o_proj": {"inputs": [[[2.0625]]], "outputs": [[[2.703125]], [[3.96875]]], "params": {"weight": [[0.28125]]}}, "model.layers.77.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.84375]], [[15.9375]]]}, "model.layers.77.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.625]]]}, "model.layers.77.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[36.0]]]}, "model.layers.77.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.75]]]}, "model.layers.77.self_attn.attn.impl.k_cache": {"inputs": [[[15.9375]]]}, "model.layers.77.self_attn.attn.impl.v_cache": {"inputs": [[[2.625]]]}, "model.layers.77.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[12.5625]], [[13.0625]], [[2.25]]], "outputs": [[[1.3828125]], [[1.0]]]}, "model.layers.77.mlp.gate_up_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.265625]]}}, "model.layers.77.mlp.down_proj": {"inputs": [[[12.1875]]], "outputs": [[[4.03125]], [[6.8125]]], "params": {"weight": [[0.50390625]]}}, "model.layers.78.self_attn.qkv_proj": {"inputs": [[[8.625]]], "params": {"weight": [[0.263671875]]}}, "model.layers.78.self_attn.o_proj": {"inputs": [[[3.0625]]], "outputs": [[[2.75]], [[4.03125]]], "params": {"weight": [[0.173828125]]}}, "model.layers.78.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.87109375]], [[17.625]]]}, "model.layers.78.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.65625]]]}, "model.layers.78.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[30.5]]]}, "model.layers.78.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.6875]]]}, "model.layers.78.self_attn.attn.impl.k_cache": {"inputs": [[[17.625]]]}, "model.layers.78.self_attn.attn.impl.v_cache": {"inputs": [[[3.65625]]]}, "model.layers.78.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[10.875]], [[15.5]], [[3.03125]]], "outputs": [[[2.28125]], [[1.0]]]}, "model.layers.78.mlp.gate_up_proj": {"inputs": [[[14.0]]], "params": {"weight": [[0.40625]]}}, "model.layers.78.mlp.down_proj": {"inputs": [[[73.0]]], "outputs": [[[86.5]], [[125.0]]], "params": {"weight": [[0.546875]]}}, "model.layers.79.self_attn.qkv_proj": {"inputs": [[[9.375]]], "params": {"weight": [[0.31640625]]}}, "model.layers.79.self_attn.o_proj": {"inputs": [[[3.484375]]], "outputs": [[[4.125]], [[5.15625]]], "params": {"weight": [[0.318359375]]}}, "model.layers.79.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.59375]], [[11.9375]]]}, "model.layers.79.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.28125]]]}, "model.layers.79.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[17.625]]]}, "model.layers.79.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[11.75]]]}, "model.layers.79.self_attn.attn.impl.k_cache": {"inputs": [[[11.9375]]]}, "model.layers.79.self_attn.attn.impl.v_cache": {"inputs": [[[4.28125]]]}, "model.layers.79.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.8125]], [[10.3125]], [[3.46875]]], "outputs": [[[2.109375]], [[1.0]]]}, "model.layers.79.mlp.gate_up_proj": {"inputs": [[[17.25]]], "params": {"weight": [[0.390625]]}}, "model.layers.79.mlp.down_proj": {"inputs": [[[47.5]]], "outputs": [[[155.0]], [[402.0]]], "params": {"weight": [[0.451171875]]}}, "lm_head": {"inputs": [[[81.0]]], "params": {"weight": [[0.294921875]]}}}}

quant/g3/inc_output_hooks_maxabs_2_4.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cbd1dcfe8c3012610aa934c544fc532e85858aa306a3177b9cbed9ae56e5fdff
+size 206298

quant/g3/inc_output_hooks_maxabs_2_4_mod_list.json

@@ -0,0 +1,963 @@
+[
+    "model.layers.0.self_attn.qkv_proj",
+    "model.layers.0.self_attn.o_proj",
+    "model.layers.0.self_attn.attn.impl.matmul_qk",
+    "model.layers.0.self_attn.attn.impl.softmax",
+    "model.layers.0.self_attn.attn.impl.matmul_av",
+    "model.layers.0.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.0.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.0.self_attn.attn.impl.k_cache",
+    "model.layers.0.self_attn.attn.impl.v_cache",
+    "model.layers.0.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.0.mlp.gate_up_proj",
+    "model.layers.0.mlp.down_proj",
+    "model.layers.1.self_attn.qkv_proj",
+    "model.layers.1.self_attn.o_proj",
+    "model.layers.1.self_attn.attn.impl.matmul_qk",
+    "model.layers.1.self_attn.attn.impl.softmax",
+    "model.layers.1.self_attn.attn.impl.matmul_av",
+    "model.layers.1.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.1.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.1.self_attn.attn.impl.k_cache",
+    "model.layers.1.self_attn.attn.impl.v_cache",
+    "model.layers.1.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.1.mlp.gate_up_proj",
+    "model.layers.1.mlp.down_proj",
+    "model.layers.2.self_attn.qkv_proj",
+    "model.layers.2.self_attn.o_proj",
+    "model.layers.2.self_attn.attn.impl.matmul_qk",
+    "model.layers.2.self_attn.attn.impl.softmax",
+    "model.layers.2.self_attn.attn.impl.matmul_av",
+    "model.layers.2.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.2.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.2.self_attn.attn.impl.k_cache",
+    "model.layers.2.self_attn.attn.impl.v_cache",
+    "model.layers.2.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.2.mlp.gate_up_proj",
+    "model.layers.2.mlp.down_proj",
+    "model.layers.3.self_attn.qkv_proj",
+    "model.layers.3.self_attn.o_proj",
+    "model.layers.3.self_attn.attn.impl.matmul_qk",
+    "model.layers.3.self_attn.attn.impl.softmax",
+    "model.layers.3.self_attn.attn.impl.matmul_av",
+    "model.layers.3.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.3.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.3.self_attn.attn.impl.k_cache",
+    "model.layers.3.self_attn.attn.impl.v_cache",
+    "model.layers.3.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.3.mlp.gate_up_proj",
+    "model.layers.3.mlp.down_proj",
+    "model.layers.4.self_attn.qkv_proj",
+    "model.layers.4.self_attn.o_proj",
+    "model.layers.4.self_attn.attn.impl.matmul_qk",
+    "model.layers.4.self_attn.attn.impl.softmax",
+    "model.layers.4.self_attn.attn.impl.matmul_av",
+    "model.layers.4.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.4.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.4.self_attn.attn.impl.k_cache",
+    "model.layers.4.self_attn.attn.impl.v_cache",
+    "model.layers.4.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.4.mlp.gate_up_proj",
+    "model.layers.4.mlp.down_proj",
+    "model.layers.5.self_attn.qkv_proj",
+    "model.layers.5.self_attn.o_proj",
+    "model.layers.5.self_attn.attn.impl.matmul_qk",
+    "model.layers.5.self_attn.attn.impl.softmax",
+    "model.layers.5.self_attn.attn.impl.matmul_av",
+    "model.layers.5.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.5.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.5.self_attn.attn.impl.k_cache",
+    "model.layers.5.self_attn.attn.impl.v_cache",
+    "model.layers.5.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.5.mlp.gate_up_proj",
+    "model.layers.5.mlp.down_proj",
+    "model.layers.6.self_attn.qkv_proj",
+    "model.layers.6.self_attn.o_proj",
+    "model.layers.6.self_attn.attn.impl.matmul_qk",
+    "model.layers.6.self_attn.attn.impl.softmax",
+    "model.layers.6.self_attn.attn.impl.matmul_av",
+    "model.layers.6.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.6.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.6.self_attn.attn.impl.k_cache",
+    "model.layers.6.self_attn.attn.impl.v_cache",
+    "model.layers.6.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.6.mlp.gate_up_proj",
+    "model.layers.6.mlp.down_proj",
+    "model.layers.7.self_attn.qkv_proj",
+    "model.layers.7.self_attn.o_proj",
+    "model.layers.7.self_attn.attn.impl.matmul_qk",
+    "model.layers.7.self_attn.attn.impl.softmax",
+    "model.layers.7.self_attn.attn.impl.matmul_av",
+    "model.layers.7.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.7.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.7.self_attn.attn.impl.k_cache",
+    "model.layers.7.self_attn.attn.impl.v_cache",
+    "model.layers.7.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.7.mlp.gate_up_proj",
+    "model.layers.7.mlp.down_proj",
+    "model.layers.8.self_attn.qkv_proj",
+    "model.layers.8.self_attn.o_proj",
+    "model.layers.8.self_attn.attn.impl.matmul_qk",
+    "model.layers.8.self_attn.attn.impl.softmax",
+    "model.layers.8.self_attn.attn.impl.matmul_av",
+    "model.layers.8.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.8.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.8.self_attn.attn.impl.k_cache",
+    "model.layers.8.self_attn.attn.impl.v_cache",
+    "model.layers.8.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.8.mlp.gate_up_proj",
+    "model.layers.8.mlp.down_proj",
+    "model.layers.9.self_attn.qkv_proj",
+    "model.layers.9.self_attn.o_proj",
+    "model.layers.9.self_attn.attn.impl.matmul_qk",
+    "model.layers.9.self_attn.attn.impl.softmax",
+    "model.layers.9.self_attn.attn.impl.matmul_av",
+    "model.layers.9.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.9.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.9.self_attn.attn.impl.k_cache",
+    "model.layers.9.self_attn.attn.impl.v_cache",
+    "model.layers.9.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.9.mlp.gate_up_proj",
+    "model.layers.9.mlp.down_proj",
+    "model.layers.10.self_attn.qkv_proj",
+    "model.layers.10.self_attn.o_proj",
+    "model.layers.10.self_attn.attn.impl.matmul_qk",
+    "model.layers.10.self_attn.attn.impl.softmax",
+    "model.layers.10.self_attn.attn.impl.matmul_av",
+    "model.layers.10.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.10.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.10.self_attn.attn.impl.k_cache",
+    "model.layers.10.self_attn.attn.impl.v_cache",
+    "model.layers.10.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.10.mlp.gate_up_proj",
+    "model.layers.10.mlp.down_proj",
+    "model.layers.11.self_attn.qkv_proj",
+    "model.layers.11.self_attn.o_proj",
+    "model.layers.11.self_attn.attn.impl.matmul_qk",
+    "model.layers.11.self_attn.attn.impl.softmax",
+    "model.layers.11.self_attn.attn.impl.matmul_av",
+    "model.layers.11.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.11.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.11.self_attn.attn.impl.k_cache",
+    "model.layers.11.self_attn.attn.impl.v_cache",
+    "model.layers.11.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.11.mlp.gate_up_proj",
+    "model.layers.11.mlp.down_proj",
+    "model.layers.12.self_attn.qkv_proj",
+    "model.layers.12.self_attn.o_proj",
+    "model.layers.12.self_attn.attn.impl.matmul_qk",
+    "model.layers.12.self_attn.attn.impl.softmax",
+    "model.layers.12.self_attn.attn.impl.matmul_av",
+    "model.layers.12.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.12.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.12.self_attn.attn.impl.k_cache",
+    "model.layers.12.self_attn.attn.impl.v_cache",
+    "model.layers.12.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.12.mlp.gate_up_proj",
+    "model.layers.12.mlp.down_proj",
+    "model.layers.13.self_attn.qkv_proj",
+    "model.layers.13.self_attn.o_proj",
+    "model.layers.13.self_attn.attn.impl.matmul_qk",
+    "model.layers.13.self_attn.attn.impl.softmax",
+    "model.layers.13.self_attn.attn.impl.matmul_av",
+    "model.layers.13.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.13.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.13.self_attn.attn.impl.k_cache",
+    "model.layers.13.self_attn.attn.impl.v_cache",
+    "model.layers.13.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.13.mlp.gate_up_proj",
+    "model.layers.13.mlp.down_proj",
+    "model.layers.14.self_attn.qkv_proj",
+    "model.layers.14.self_attn.o_proj",
+    "model.layers.14.self_attn.attn.impl.matmul_qk",
+    "model.layers.14.self_attn.attn.impl.softmax",
+    "model.layers.14.self_attn.attn.impl.matmul_av",
+    "model.layers.14.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.14.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.14.self_attn.attn.impl.k_cache",
+    "model.layers.14.self_attn.attn.impl.v_cache",
+    "model.layers.14.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.14.mlp.gate_up_proj",
+    "model.layers.14.mlp.down_proj",
+    "model.layers.15.self_attn.qkv_proj",
+    "model.layers.15.self_attn.o_proj",
+    "model.layers.15.self_attn.attn.impl.matmul_qk",
+    "model.layers.15.self_attn.attn.impl.softmax",
+    "model.layers.15.self_attn.attn.impl.matmul_av",
+    "model.layers.15.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.15.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.15.self_attn.attn.impl.k_cache",
+    "model.layers.15.self_attn.attn.impl.v_cache",
+    "model.layers.15.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.15.mlp.gate_up_proj",
+    "model.layers.15.mlp.down_proj",
+    "model.layers.16.self_attn.qkv_proj",
+    "model.layers.16.self_attn.o_proj",
+    "model.layers.16.self_attn.attn.impl.matmul_qk",
+    "model.layers.16.self_attn.attn.impl.softmax",
+    "model.layers.16.self_attn.attn.impl.matmul_av",
+    "model.layers.16.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.16.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.16.self_attn.attn.impl.k_cache",
+    "model.layers.16.self_attn.attn.impl.v_cache",
+    "model.layers.16.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.16.mlp.gate_up_proj",
+    "model.layers.16.mlp.down_proj",
+    "model.layers.17.self_attn.qkv_proj",
+    "model.layers.17.self_attn.o_proj",
+    "model.layers.17.self_attn.attn.impl.matmul_qk",
+    "model.layers.17.self_attn.attn.impl.softmax",
+    "model.layers.17.self_attn.attn.impl.matmul_av",
+    "model.layers.17.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.17.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.17.self_attn.attn.impl.k_cache",
+    "model.layers.17.self_attn.attn.impl.v_cache",
+    "model.layers.17.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.17.mlp.gate_up_proj",
+    "model.layers.17.mlp.down_proj",
+    "model.layers.18.self_attn.qkv_proj",
+    "model.layers.18.self_attn.o_proj",
+    "model.layers.18.self_attn.attn.impl.matmul_qk",
+    "model.layers.18.self_attn.attn.impl.softmax",
+    "model.layers.18.self_attn.attn.impl.matmul_av",
+    "model.layers.18.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.18.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.18.self_attn.attn.impl.k_cache",
+    "model.layers.18.self_attn.attn.impl.v_cache",
+    "model.layers.18.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.18.mlp.gate_up_proj",
+    "model.layers.18.mlp.down_proj",
+    "model.layers.19.self_attn.qkv_proj",
+    "model.layers.19.self_attn.o_proj",
+    "model.layers.19.self_attn.attn.impl.matmul_qk",
+    "model.layers.19.self_attn.attn.impl.softmax",
+    "model.layers.19.self_attn.attn.impl.matmul_av",
+    "model.layers.19.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.19.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.19.self_attn.attn.impl.k_cache",
+    "model.layers.19.self_attn.attn.impl.v_cache",
+    "model.layers.19.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.19.mlp.gate_up_proj",
+    "model.layers.19.mlp.down_proj",
+    "model.layers.20.self_attn.qkv_proj",
+    "model.layers.20.self_attn.o_proj",
+    "model.layers.20.self_attn.attn.impl.matmul_qk",
+    "model.layers.20.self_attn.attn.impl.softmax",
+    "model.layers.20.self_attn.attn.impl.matmul_av",
+    "model.layers.20.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.20.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.20.self_attn.attn.impl.k_cache",
+    "model.layers.20.self_attn.attn.impl.v_cache",
+    "model.layers.20.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.20.mlp.gate_up_proj",
+    "model.layers.20.mlp.down_proj",
+    "model.layers.21.self_attn.qkv_proj",
+    "model.layers.21.self_attn.o_proj",
+    "model.layers.21.self_attn.attn.impl.matmul_qk",
+    "model.layers.21.self_attn.attn.impl.softmax",
+    "model.layers.21.self_attn.attn.impl.matmul_av",
+    "model.layers.21.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.21.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.21.self_attn.attn.impl.k_cache",
+    "model.layers.21.self_attn.attn.impl.v_cache",
+    "model.layers.21.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.21.mlp.gate_up_proj",
+    "model.layers.21.mlp.down_proj",
+    "model.layers.22.self_attn.qkv_proj",
+    "model.layers.22.self_attn.o_proj",
+    "model.layers.22.self_attn.attn.impl.matmul_qk",
+    "model.layers.22.self_attn.attn.impl.softmax",
+    "model.layers.22.self_attn.attn.impl.matmul_av",
+    "model.layers.22.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.22.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.22.self_attn.attn.impl.k_cache",
+    "model.layers.22.self_attn.attn.impl.v_cache",
+    "model.layers.22.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.22.mlp.gate_up_proj",
+    "model.layers.22.mlp.down_proj",
+    "model.layers.23.self_attn.qkv_proj",
+    "model.layers.23.self_attn.o_proj",
+    "model.layers.23.self_attn.attn.impl.matmul_qk",
+    "model.layers.23.self_attn.attn.impl.softmax",
+    "model.layers.23.self_attn.attn.impl.matmul_av",
+    "model.layers.23.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.23.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.23.self_attn.attn.impl.k_cache",
+    "model.layers.23.self_attn.attn.impl.v_cache",
+    "model.layers.23.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.23.mlp.gate_up_proj",
+    "model.layers.23.mlp.down_proj",
+    "model.layers.24.self_attn.qkv_proj",
+    "model.layers.24.self_attn.o_proj",
+    "model.layers.24.self_attn.attn.impl.matmul_qk",
+    "model.layers.24.self_attn.attn.impl.softmax",
+    "model.layers.24.self_attn.attn.impl.matmul_av",
+    "model.layers.24.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.24.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.24.self_attn.attn.impl.k_cache",
+    "model.layers.24.self_attn.attn.impl.v_cache",
+    "model.layers.24.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.24.mlp.gate_up_proj",
+    "model.layers.24.mlp.down_proj",
+    "model.layers.25.self_attn.qkv_proj",
+    "model.layers.25.self_attn.o_proj",
+    "model.layers.25.self_attn.attn.impl.matmul_qk",
+    "model.layers.25.self_attn.attn.impl.softmax",
+    "model.layers.25.self_attn.attn.impl.matmul_av",
+    "model.layers.25.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.25.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.25.self_attn.attn.impl.k_cache",
+    "model.layers.25.self_attn.attn.impl.v_cache",
+    "model.layers.25.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.25.mlp.gate_up_proj",
+    "model.layers.25.mlp.down_proj",
+    "model.layers.26.self_attn.qkv_proj",
+    "model.layers.26.self_attn.o_proj",
+    "model.layers.26.self_attn.attn.impl.matmul_qk",
+    "model.layers.26.self_attn.attn.impl.softmax",
+    "model.layers.26.self_attn.attn.impl.matmul_av",
+    "model.layers.26.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.26.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.26.self_attn.attn.impl.k_cache",
+    "model.layers.26.self_attn.attn.impl.v_cache",
+    "model.layers.26.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.26.mlp.gate_up_proj",
+    "model.layers.26.mlp.down_proj",
+    "model.layers.27.self_attn.qkv_proj",
+    "model.layers.27.self_attn.o_proj",
+    "model.layers.27.self_attn.attn.impl.matmul_qk",
+    "model.layers.27.self_attn.attn.impl.softmax",
+    "model.layers.27.self_attn.attn.impl.matmul_av",
+    "model.layers.27.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.27.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.27.self_attn.attn.impl.k_cache",
+    "model.layers.27.self_attn.attn.impl.v_cache",
+    "model.layers.27.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.27.mlp.gate_up_proj",
+    "model.layers.27.mlp.down_proj",
+    "model.layers.28.self_attn.qkv_proj",
+    "model.layers.28.self_attn.o_proj",
+    "model.layers.28.self_attn.attn.impl.matmul_qk",
+    "model.layers.28.self_attn.attn.impl.softmax",
+    "model.layers.28.self_attn.attn.impl.matmul_av",
+    "model.layers.28.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.28.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.28.self_attn.attn.impl.k_cache",
+    "model.layers.28.self_attn.attn.impl.v_cache",
+    "model.layers.28.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.28.mlp.gate_up_proj",
+    "model.layers.28.mlp.down_proj",
+    "model.layers.29.self_attn.qkv_proj",
+    "model.layers.29.self_attn.o_proj",
+    "model.layers.29.self_attn.attn.impl.matmul_qk",
+    "model.layers.29.self_attn.attn.impl.softmax",
+    "model.layers.29.self_attn.attn.impl.matmul_av",
+    "model.layers.29.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.29.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.29.self_attn.attn.impl.k_cache",
+    "model.layers.29.self_attn.attn.impl.v_cache",
+    "model.layers.29.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.29.mlp.gate_up_proj",
+    "model.layers.29.mlp.down_proj",
+    "model.layers.30.self_attn.qkv_proj",
+    "model.layers.30.self_attn.o_proj",
+    "model.layers.30.self_attn.attn.impl.matmul_qk",
+    "model.layers.30.self_attn.attn.impl.softmax",
+    "model.layers.30.self_attn.attn.impl.matmul_av",
+    "model.layers.30.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.30.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.30.self_attn.attn.impl.k_cache",
+    "model.layers.30.self_attn.attn.impl.v_cache",
+    "model.layers.30.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.30.mlp.gate_up_proj",
+    "model.layers.30.mlp.down_proj",
+    "model.layers.31.self_attn.qkv_proj",
+    "model.layers.31.self_attn.o_proj",
+    "model.layers.31.self_attn.attn.impl.matmul_qk",
+    "model.layers.31.self_attn.attn.impl.softmax",
+    "model.layers.31.self_attn.attn.impl.matmul_av",
+    "model.layers.31.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.31.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.31.self_attn.attn.impl.k_cache",
+    "model.layers.31.self_attn.attn.impl.v_cache",
+    "model.layers.31.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.31.mlp.gate_up_proj",
+    "model.layers.31.mlp.down_proj",
+    "model.layers.32.self_attn.qkv_proj",
+    "model.layers.32.self_attn.o_proj",
+    "model.layers.32.self_attn.attn.impl.matmul_qk",
+    "model.layers.32.self_attn.attn.impl.softmax",
+    "model.layers.32.self_attn.attn.impl.matmul_av",
+    "model.layers.32.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.32.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.32.self_attn.attn.impl.k_cache",
+    "model.layers.32.self_attn.attn.impl.v_cache",
+    "model.layers.32.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.32.mlp.gate_up_proj",
+    "model.layers.32.mlp.down_proj",
+    "model.layers.33.self_attn.qkv_proj",
+    "model.layers.33.self_attn.o_proj",
+    "model.layers.33.self_attn.attn.impl.matmul_qk",
+    "model.layers.33.self_attn.attn.impl.softmax",
+    "model.layers.33.self_attn.attn.impl.matmul_av",
+    "model.layers.33.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.33.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.33.self_attn.attn.impl.k_cache",
+    "model.layers.33.self_attn.attn.impl.v_cache",
+    "model.layers.33.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.33.mlp.gate_up_proj",
+    "model.layers.33.mlp.down_proj",
+    "model.layers.34.self_attn.qkv_proj",
+    "model.layers.34.self_attn.o_proj",
+    "model.layers.34.self_attn.attn.impl.matmul_qk",
+    "model.layers.34.self_attn.attn.impl.softmax",
+    "model.layers.34.self_attn.attn.impl.matmul_av",
+    "model.layers.34.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.34.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.34.self_attn.attn.impl.k_cache",
+    "model.layers.34.self_attn.attn.impl.v_cache",
+    "model.layers.34.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.34.mlp.gate_up_proj",
+    "model.layers.34.mlp.down_proj",
+    "model.layers.35.self_attn.qkv_proj",
+    "model.layers.35.self_attn.o_proj",
+    "model.layers.35.self_attn.attn.impl.matmul_qk",
+    "model.layers.35.self_attn.attn.impl.softmax",
+    "model.layers.35.self_attn.attn.impl.matmul_av",
+    "model.layers.35.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.35.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.35.self_attn.attn.impl.k_cache",
+    "model.layers.35.self_attn.attn.impl.v_cache",
+    "model.layers.35.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.35.mlp.gate_up_proj",
+    "model.layers.35.mlp.down_proj",
+    "model.layers.36.self_attn.qkv_proj",
+    "model.layers.36.self_attn.o_proj",
+    "model.layers.36.self_attn.attn.impl.matmul_qk",
+    "model.layers.36.self_attn.attn.impl.softmax",
+    "model.layers.36.self_attn.attn.impl.matmul_av",
+    "model.layers.36.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.36.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.36.self_attn.attn.impl.k_cache",
+    "model.layers.36.self_attn.attn.impl.v_cache",
+    "model.layers.36.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.36.mlp.gate_up_proj",
+    "model.layers.36.mlp.down_proj",
+    "model.layers.37.self_attn.qkv_proj",
+    "model.layers.37.self_attn.o_proj",
+    "model.layers.37.self_attn.attn.impl.matmul_qk",
+    "model.layers.37.self_attn.attn.impl.softmax",
+    "model.layers.37.self_attn.attn.impl.matmul_av",
+    "model.layers.37.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.37.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.37.self_attn.attn.impl.k_cache",
+    "model.layers.37.self_attn.attn.impl.v_cache",
+    "model.layers.37.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.37.mlp.gate_up_proj",
+    "model.layers.37.mlp.down_proj",
+    "model.layers.38.self_attn.qkv_proj",
+    "model.layers.38.self_attn.o_proj",
+    "model.layers.38.self_attn.attn.impl.matmul_qk",
+    "model.layers.38.self_attn.attn.impl.softmax",
+    "model.layers.38.self_attn.attn.impl.matmul_av",
+    "model.layers.38.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.38.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.38.self_attn.attn.impl.k_cache",
+    "model.layers.38.self_attn.attn.impl.v_cache",
+    "model.layers.38.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.38.mlp.gate_up_proj",
+    "model.layers.38.mlp.down_proj",
+    "model.layers.39.self_attn.qkv_proj",
+    "model.layers.39.self_attn.o_proj",
+    "model.layers.39.self_attn.attn.impl.matmul_qk",
+    "model.layers.39.self_attn.attn.impl.softmax",
+    "model.layers.39.self_attn.attn.impl.matmul_av",
+    "model.layers.39.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.39.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.39.self_attn.attn.impl.k_cache",
+    "model.layers.39.self_attn.attn.impl.v_cache",
+    "model.layers.39.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.39.mlp.gate_up_proj",
+    "model.layers.39.mlp.down_proj",
+    "model.layers.40.self_attn.qkv_proj",
+    "model.layers.40.self_attn.o_proj",
+    "model.layers.40.self_attn.attn.impl.matmul_qk",
+    "model.layers.40.self_attn.attn.impl.softmax",
+    "model.layers.40.self_attn.attn.impl.matmul_av",
+    "model.layers.40.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.40.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.40.self_attn.attn.impl.k_cache",
+    "model.layers.40.self_attn.attn.impl.v_cache",
+    "model.layers.40.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.40.mlp.gate_up_proj",
+    "model.layers.40.mlp.down_proj",
+    "model.layers.41.self_attn.qkv_proj",
+    "model.layers.41.self_attn.o_proj",
+    "model.layers.41.self_attn.attn.impl.matmul_qk",
+    "model.layers.41.self_attn.attn.impl.softmax",
+    "model.layers.41.self_attn.attn.impl.matmul_av",
+    "model.layers.41.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.41.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.41.self_attn.attn.impl.k_cache",
+    "model.layers.41.self_attn.attn.impl.v_cache",
+    "model.layers.41.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.41.mlp.gate_up_proj",
+    "model.layers.41.mlp.down_proj",
+    "model.layers.42.self_attn.qkv_proj",
+    "model.layers.42.self_attn.o_proj",
+    "model.layers.42.self_attn.attn.impl.matmul_qk",
+    "model.layers.42.self_attn.attn.impl.softmax",
+    "model.layers.42.self_attn.attn.impl.matmul_av",
+    "model.layers.42.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.42.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.42.self_attn.attn.impl.k_cache",
+    "model.layers.42.self_attn.attn.impl.v_cache",
+    "model.layers.42.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.42.mlp.gate_up_proj",
+    "model.layers.42.mlp.down_proj",
+    "model.layers.43.self_attn.qkv_proj",
+    "model.layers.43.self_attn.o_proj",
+    "model.layers.43.self_attn.attn.impl.matmul_qk",
+    "model.layers.43.self_attn.attn.impl.softmax",
+    "model.layers.43.self_attn.attn.impl.matmul_av",
+    "model.layers.43.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.43.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.43.self_attn.attn.impl.k_cache",
+    "model.layers.43.self_attn.attn.impl.v_cache",
+    "model.layers.43.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.43.mlp.gate_up_proj",
+    "model.layers.43.mlp.down_proj",
+    "model.layers.44.self_attn.qkv_proj",
+    "model.layers.44.self_attn.o_proj",
+    "model.layers.44.self_attn.attn.impl.matmul_qk",
+    "model.layers.44.self_attn.attn.impl.softmax",
+    "model.layers.44.self_attn.attn.impl.matmul_av",
+    "model.layers.44.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.44.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.44.self_attn.attn.impl.k_cache",
+    "model.layers.44.self_attn.attn.impl.v_cache",
+    "model.layers.44.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.44.mlp.gate_up_proj",
+    "model.layers.44.mlp.down_proj",
+    "model.layers.45.self_attn.qkv_proj",
+    "model.layers.45.self_attn.o_proj",
+    "model.layers.45.self_attn.attn.impl.matmul_qk",
+    "model.layers.45.self_attn.attn.impl.softmax",
+    "model.layers.45.self_attn.attn.impl.matmul_av",
+    "model.layers.45.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.45.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.45.self_attn.attn.impl.k_cache",
+    "model.layers.45.self_attn.attn.impl.v_cache",
+    "model.layers.45.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.45.mlp.gate_up_proj",
+    "model.layers.45.mlp.down_proj",
+    "model.layers.46.self_attn.qkv_proj",
+    "model.layers.46.self_attn.o_proj",
+    "model.layers.46.self_attn.attn.impl.matmul_qk",
+    "model.layers.46.self_attn.attn.impl.softmax",
+    "model.layers.46.self_attn.attn.impl.matmul_av",
+    "model.layers.46.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.46.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.46.self_attn.attn.impl.k_cache",
+    "model.layers.46.self_attn.attn.impl.v_cache",
+    "model.layers.46.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.46.mlp.gate_up_proj",
+    "model.layers.46.mlp.down_proj",
+    "model.layers.47.self_attn.qkv_proj",
+    "model.layers.47.self_attn.o_proj",
+    "model.layers.47.self_attn.attn.impl.matmul_qk",
+    "model.layers.47.self_attn.attn.impl.softmax",
+    "model.layers.47.self_attn.attn.impl.matmul_av",
+    "model.layers.47.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.47.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.47.self_attn.attn.impl.k_cache",
+    "model.layers.47.self_attn.attn.impl.v_cache",
+    "model.layers.47.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.47.mlp.gate_up_proj",
+    "model.layers.47.mlp.down_proj",
+    "model.layers.48.self_attn.qkv_proj",
+    "model.layers.48.self_attn.o_proj",
+    "model.layers.48.self_attn.attn.impl.matmul_qk",
+    "model.layers.48.self_attn.attn.impl.softmax",
+    "model.layers.48.self_attn.attn.impl.matmul_av",
+    "model.layers.48.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.48.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.48.self_attn.attn.impl.k_cache",
+    "model.layers.48.self_attn.attn.impl.v_cache",
+    "model.layers.48.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.48.mlp.gate_up_proj",
+    "model.layers.48.mlp.down_proj",
+    "model.layers.49.self_attn.qkv_proj",
+    "model.layers.49.self_attn.o_proj",
+    "model.layers.49.self_attn.attn.impl.matmul_qk",
+    "model.layers.49.self_attn.attn.impl.softmax",
+    "model.layers.49.self_attn.attn.impl.matmul_av",
+    "model.layers.49.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.49.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.49.self_attn.attn.impl.k_cache",
+    "model.layers.49.self_attn.attn.impl.v_cache",
+    "model.layers.49.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.49.mlp.gate_up_proj",
+    "model.layers.49.mlp.down_proj",
+    "model.layers.50.self_attn.qkv_proj",
+    "model.layers.50.self_attn.o_proj",
+    "model.layers.50.self_attn.attn.impl.matmul_qk",
+    "model.layers.50.self_attn.attn.impl.softmax",
+    "model.layers.50.self_attn.attn.impl.matmul_av",
+    "model.layers.50.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.50.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.50.self_attn.attn.impl.k_cache",
+    "model.layers.50.self_attn.attn.impl.v_cache",
+    "model.layers.50.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.50.mlp.gate_up_proj",
+    "model.layers.50.mlp.down_proj",
+    "model.layers.51.self_attn.qkv_proj",
+    "model.layers.51.self_attn.o_proj",
+    "model.layers.51.self_attn.attn.impl.matmul_qk",
+    "model.layers.51.self_attn.attn.impl.softmax",
+    "model.layers.51.self_attn.attn.impl.matmul_av",
+    "model.layers.51.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.51.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.51.self_attn.attn.impl.k_cache",
+    "model.layers.51.self_attn.attn.impl.v_cache",
+    "model.layers.51.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.51.mlp.gate_up_proj",
+    "model.layers.51.mlp.down_proj",
+    "model.layers.52.self_attn.qkv_proj",
+    "model.layers.52.self_attn.o_proj",
+    "model.layers.52.self_attn.attn.impl.matmul_qk",
+    "model.layers.52.self_attn.attn.impl.softmax",
+    "model.layers.52.self_attn.attn.impl.matmul_av",
+    "model.layers.52.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.52.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.52.self_attn.attn.impl.k_cache",
+    "model.layers.52.self_attn.attn.impl.v_cache",
+    "model.layers.52.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.52.mlp.gate_up_proj",
+    "model.layers.52.mlp.down_proj",
+    "model.layers.53.self_attn.qkv_proj",
+    "model.layers.53.self_attn.o_proj",
+    "model.layers.53.self_attn.attn.impl.matmul_qk",
+    "model.layers.53.self_attn.attn.impl.softmax",
+    "model.layers.53.self_attn.attn.impl.matmul_av",
+    "model.layers.53.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.53.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.53.self_attn.attn.impl.k_cache",
+    "model.layers.53.self_attn.attn.impl.v_cache",
+    "model.layers.53.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.53.mlp.gate_up_proj",
+    "model.layers.53.mlp.down_proj",
+    "model.layers.54.self_attn.qkv_proj",
+    "model.layers.54.self_attn.o_proj",
+    "model.layers.54.self_attn.attn.impl.matmul_qk",
+    "model.layers.54.self_attn.attn.impl.softmax",
+    "model.layers.54.self_attn.attn.impl.matmul_av",
+    "model.layers.54.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.54.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.54.self_attn.attn.impl.k_cache",
+    "model.layers.54.self_attn.attn.impl.v_cache",
+    "model.layers.54.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.54.mlp.gate_up_proj",
+    "model.layers.54.mlp.down_proj",
+    "model.layers.55.self_attn.qkv_proj",
+    "model.layers.55.self_attn.o_proj",
+    "model.layers.55.self_attn.attn.impl.matmul_qk",
+    "model.layers.55.self_attn.attn.impl.softmax",
+    "model.layers.55.self_attn.attn.impl.matmul_av",
+    "model.layers.55.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.55.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.55.self_attn.attn.impl.k_cache",
+    "model.layers.55.self_attn.attn.impl.v_cache",
+    "model.layers.55.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.55.mlp.gate_up_proj",
+    "model.layers.55.mlp.down_proj",
+    "model.layers.56.self_attn.qkv_proj",
+    "model.layers.56.self_attn.o_proj",
+    "model.layers.56.self_attn.attn.impl.matmul_qk",
+    "model.layers.56.self_attn.attn.impl.softmax",
+    "model.layers.56.self_attn.attn.impl.matmul_av",
+    "model.layers.56.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.56.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.56.self_attn.attn.impl.k_cache",
+    "model.layers.56.self_attn.attn.impl.v_cache",
+    "model.layers.56.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.56.mlp.gate_up_proj",
+    "model.layers.56.mlp.down_proj",
+    "model.layers.57.self_attn.qkv_proj",
+    "model.layers.57.self_attn.o_proj",
+    "model.layers.57.self_attn.attn.impl.matmul_qk",
+    "model.layers.57.self_attn.attn.impl.softmax",
+    "model.layers.57.self_attn.attn.impl.matmul_av",
+    "model.layers.57.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.57.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.57.self_attn.attn.impl.k_cache",
+    "model.layers.57.self_attn.attn.impl.v_cache",
+    "model.layers.57.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.57.mlp.gate_up_proj",
+    "model.layers.57.mlp.down_proj",
+    "model.layers.58.self_attn.qkv_proj",
+    "model.layers.58.self_attn.o_proj",
+    "model.layers.58.self_attn.attn.impl.matmul_qk",
+    "model.layers.58.self_attn.attn.impl.softmax",
+    "model.layers.58.self_attn.attn.impl.matmul_av",
+    "model.layers.58.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.58.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.58.self_attn.attn.impl.k_cache",
+    "model.layers.58.self_attn.attn.impl.v_cache",
+    "model.layers.58.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.58.mlp.gate_up_proj",
+    "model.layers.58.mlp.down_proj",
+    "model.layers.59.self_attn.qkv_proj",
+    "model.layers.59.self_attn.o_proj",
+    "model.layers.59.self_attn.attn.impl.matmul_qk",
+    "model.layers.59.self_attn.attn.impl.softmax",
+    "model.layers.59.self_attn.attn.impl.matmul_av",
+    "model.layers.59.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.59.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.59.self_attn.attn.impl.k_cache",
+    "model.layers.59.self_attn.attn.impl.v_cache",
+    "model.layers.59.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.59.mlp.gate_up_proj",
+    "model.layers.59.mlp.down_proj",
+    "model.layers.60.self_attn.qkv_proj",
+    "model.layers.60.self_attn.o_proj",
+    "model.layers.60.self_attn.attn.impl.matmul_qk",
+    "model.layers.60.self_attn.attn.impl.softmax",
+    "model.layers.60.self_attn.attn.impl.matmul_av",
+    "model.layers.60.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.60.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.60.self_attn.attn.impl.k_cache",
+    "model.layers.60.self_attn.attn.impl.v_cache",
+    "model.layers.60.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.60.mlp.gate_up_proj",
+    "model.layers.60.mlp.down_proj",
+    "model.layers.61.self_attn.qkv_proj",
+    "model.layers.61.self_attn.o_proj",
+    "model.layers.61.self_attn.attn.impl.matmul_qk",
+    "model.layers.61.self_attn.attn.impl.softmax",
+    "model.layers.61.self_attn.attn.impl.matmul_av",
+    "model.layers.61.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.61.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.61.self_attn.attn.impl.k_cache",
+    "model.layers.61.self_attn.attn.impl.v_cache",
+    "model.layers.61.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.61.mlp.gate_up_proj",
+    "model.layers.61.mlp.down_proj",
+    "model.layers.62.self_attn.qkv_proj",
+    "model.layers.62.self_attn.o_proj",
+    "model.layers.62.self_attn.attn.impl.matmul_qk",
+    "model.layers.62.self_attn.attn.impl.softmax",
+    "model.layers.62.self_attn.attn.impl.matmul_av",
+    "model.layers.62.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.62.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.62.self_attn.attn.impl.k_cache",
+    "model.layers.62.self_attn.attn.impl.v_cache",
+    "model.layers.62.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.62.mlp.gate_up_proj",
+    "model.layers.62.mlp.down_proj",
+    "model.layers.63.self_attn.qkv_proj",
+    "model.layers.63.self_attn.o_proj",
+    "model.layers.63.self_attn.attn.impl.matmul_qk",
+    "model.layers.63.self_attn.attn.impl.softmax",
+    "model.layers.63.self_attn.attn.impl.matmul_av",
+    "model.layers.63.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.63.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.63.self_attn.attn.impl.k_cache",
+    "model.layers.63.self_attn.attn.impl.v_cache",
+    "model.layers.63.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.63.mlp.gate_up_proj",
+    "model.layers.63.mlp.down_proj",
+    "model.layers.64.self_attn.qkv_proj",
+    "model.layers.64.self_attn.o_proj",
+    "model.layers.64.self_attn.attn.impl.matmul_qk",
+    "model.layers.64.self_attn.attn.impl.softmax",
+    "model.layers.64.self_attn.attn.impl.matmul_av",
+    "model.layers.64.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.64.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.64.self_attn.attn.impl.k_cache",
+    "model.layers.64.self_attn.attn.impl.v_cache",
+    "model.layers.64.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.64.mlp.gate_up_proj",
+    "model.layers.64.mlp.down_proj",
+    "model.layers.65.self_attn.qkv_proj",
+    "model.layers.65.self_attn.o_proj",
+    "model.layers.65.self_attn.attn.impl.matmul_qk",
+    "model.layers.65.self_attn.attn.impl.softmax",
+    "model.layers.65.self_attn.attn.impl.matmul_av",
+    "model.layers.65.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.65.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.65.self_attn.attn.impl.k_cache",
+    "model.layers.65.self_attn.attn.impl.v_cache",
+    "model.layers.65.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.65.mlp.gate_up_proj",
+    "model.layers.65.mlp.down_proj",
+    "model.layers.66.self_attn.qkv_proj",
+    "model.layers.66.self_attn.o_proj",
+    "model.layers.66.self_attn.attn.impl.matmul_qk",
+    "model.layers.66.self_attn.attn.impl.softmax",
+    "model.layers.66.self_attn.attn.impl.matmul_av",
+    "model.layers.66.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.66.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.66.self_attn.attn.impl.k_cache",
+    "model.layers.66.self_attn.attn.impl.v_cache",
+    "model.layers.66.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.66.mlp.gate_up_proj",
+    "model.layers.66.mlp.down_proj",
+    "model.layers.67.self_attn.qkv_proj",
+    "model.layers.67.self_attn.o_proj",
+    "model.layers.67.self_attn.attn.impl.matmul_qk",
+    "model.layers.67.self_attn.attn.impl.softmax",
+    "model.layers.67.self_attn.attn.impl.matmul_av",
+    "model.layers.67.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.67.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.67.self_attn.attn.impl.k_cache",
+    "model.layers.67.self_attn.attn.impl.v_cache",
+    "model.layers.67.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.67.mlp.gate_up_proj",
+    "model.layers.67.mlp.down_proj",
+    "model.layers.68.self_attn.qkv_proj",
+    "model.layers.68.self_attn.o_proj",
+    "model.layers.68.self_attn.attn.impl.matmul_qk",
+    "model.layers.68.self_attn.attn.impl.softmax",
+    "model.layers.68.self_attn.attn.impl.matmul_av",
+    "model.layers.68.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.68.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.68.self_attn.attn.impl.k_cache",
+    "model.layers.68.self_attn.attn.impl.v_cache",
+    "model.layers.68.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.68.mlp.gate_up_proj",
+    "model.layers.68.mlp.down_proj",
+    "model.layers.69.self_attn.qkv_proj",
+    "model.layers.69.self_attn.o_proj",
+    "model.layers.69.self_attn.attn.impl.matmul_qk",
+    "model.layers.69.self_attn.attn.impl.softmax",
+    "model.layers.69.self_attn.attn.impl.matmul_av",
+    "model.layers.69.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.69.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.69.self_attn.attn.impl.k_cache",
+    "model.layers.69.self_attn.attn.impl.v_cache",
+    "model.layers.69.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.69.mlp.gate_up_proj",
+    "model.layers.69.mlp.down_proj",
+    "model.layers.70.self_attn.qkv_proj",
+    "model.layers.70.self_attn.o_proj",
+    "model.layers.70.self_attn.attn.impl.matmul_qk",
+    "model.layers.70.self_attn.attn.impl.softmax",
+    "model.layers.70.self_attn.attn.impl.matmul_av",
+    "model.layers.70.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.70.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.70.self_attn.attn.impl.k_cache",
+    "model.layers.70.self_attn.attn.impl.v_cache",
+    "model.layers.70.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.70.mlp.gate_up_proj",
+    "model.layers.70.mlp.down_proj",
+    "model.layers.71.self_attn.qkv_proj",
+    "model.layers.71.self_attn.o_proj",
+    "model.layers.71.self_attn.attn.impl.matmul_qk",
+    "model.layers.71.self_attn.attn.impl.softmax",
+    "model.layers.71.self_attn.attn.impl.matmul_av",
+    "model.layers.71.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.71.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.71.self_attn.attn.impl.k_cache",
+    "model.layers.71.self_attn.attn.impl.v_cache",
+    "model.layers.71.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.71.mlp.gate_up_proj",
+    "model.layers.71.mlp.down_proj",
+    "model.layers.72.self_attn.qkv_proj",
+    "model.layers.72.self_attn.o_proj",
+    "model.layers.72.self_attn.attn.impl.matmul_qk",
+    "model.layers.72.self_attn.attn.impl.softmax",
+    "model.layers.72.self_attn.attn.impl.matmul_av",
+    "model.layers.72.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.72.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.72.self_attn.attn.impl.k_cache",
+    "model.layers.72.self_attn.attn.impl.v_cache",
+    "model.layers.72.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.72.mlp.gate_up_proj",
+    "model.layers.72.mlp.down_proj",
+    "model.layers.73.self_attn.qkv_proj",
+    "model.layers.73.self_attn.o_proj",
+    "model.layers.73.self_attn.attn.impl.matmul_qk",
+    "model.layers.73.self_attn.attn.impl.softmax",
+    "model.layers.73.self_attn.attn.impl.matmul_av",
+    "model.layers.73.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.73.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.73.self_attn.attn.impl.k_cache",
+    "model.layers.73.self_attn.attn.impl.v_cache",
+    "model.layers.73.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.73.mlp.gate_up_proj",
+    "model.layers.73.mlp.down_proj",
+    "model.layers.74.self_attn.qkv_proj",
+    "model.layers.74.self_attn.o_proj",
+    "model.layers.74.self_attn.attn.impl.matmul_qk",
+    "model.layers.74.self_attn.attn.impl.softmax",
+    "model.layers.74.self_attn.attn.impl.matmul_av",
+    "model.layers.74.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.74.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.74.self_attn.attn.impl.k_cache",
+    "model.layers.74.self_attn.attn.impl.v_cache",
+    "model.layers.74.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.74.mlp.gate_up_proj",
+    "model.layers.74.mlp.down_proj",
+    "model.layers.75.self_attn.qkv_proj",
+    "model.layers.75.self_attn.o_proj",
+    "model.layers.75.self_attn.attn.impl.matmul_qk",
+    "model.layers.75.self_attn.attn.impl.softmax",
+    "model.layers.75.self_attn.attn.impl.matmul_av",
+    "model.layers.75.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.75.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.75.self_attn.attn.impl.k_cache",
+    "model.layers.75.self_attn.attn.impl.v_cache",
+    "model.layers.75.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.75.mlp.gate_up_proj",
+    "model.layers.75.mlp.down_proj",
+    "model.layers.76.self_attn.qkv_proj",
+    "model.layers.76.self_attn.o_proj",
+    "model.layers.76.self_attn.attn.impl.matmul_qk",
+    "model.layers.76.self_attn.attn.impl.softmax",
+    "model.layers.76.self_attn.attn.impl.matmul_av",
+    "model.layers.76.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.76.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.76.self_attn.attn.impl.k_cache",
+    "model.layers.76.self_attn.attn.impl.v_cache",
+    "model.layers.76.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.76.mlp.gate_up_proj",
+    "model.layers.76.mlp.down_proj",
+    "model.layers.77.self_attn.qkv_proj",
+    "model.layers.77.self_attn.o_proj",
+    "model.layers.77.self_attn.attn.impl.matmul_qk",
+    "model.layers.77.self_attn.attn.impl.softmax",
+    "model.layers.77.self_attn.attn.impl.matmul_av",
+    "model.layers.77.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.77.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.77.self_attn.attn.impl.k_cache",
+    "model.layers.77.self_attn.attn.impl.v_cache",
+    "model.layers.77.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.77.mlp.gate_up_proj",
+    "model.layers.77.mlp.down_proj",
+    "model.layers.78.self_attn.qkv_proj",
+    "model.layers.78.self_attn.o_proj",
+    "model.layers.78.self_attn.attn.impl.matmul_qk",
+    "model.layers.78.self_attn.attn.impl.softmax",
+    "model.layers.78.self_attn.attn.impl.matmul_av",
+    "model.layers.78.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.78.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.78.self_attn.attn.impl.k_cache",
+    "model.layers.78.self_attn.attn.impl.v_cache",
+    "model.layers.78.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.78.mlp.gate_up_proj",
+    "model.layers.78.mlp.down_proj",
+    "model.layers.79.self_attn.qkv_proj",
+    "model.layers.79.self_attn.o_proj",
+    "model.layers.79.self_attn.attn.impl.matmul_qk",
+    "model.layers.79.self_attn.attn.impl.softmax",
+    "model.layers.79.self_attn.attn.impl.matmul_av",
+    "model.layers.79.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.79.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.79.self_attn.attn.impl.k_cache",
+    "model.layers.79.self_attn.attn.impl.v_cache",
+    "model.layers.79.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.79.mlp.gate_up_proj",
+    "model.layers.79.mlp.down_proj",
+    "lm_head"
+]

quant/g3/inc_output_hooks_maxabs_3_4.json

@@ -0,0 +1 @@
+{"GlobalRank": null, "LocalRank": 3, "Mode": "DynamicRange", "Nodes": {"model.layers.0.self_attn.qkv_proj": {"inputs": [[[4.09375]]], "params": {"weight": [[29.0]]}}, "model.layers.0.self_attn.o_proj": {"inputs": [[[0.12353515625]]], "outputs": [[[0.140625]], [[0.248046875]]], "params": {"weight": [[0.30859375]]}}, "model.layers.0.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.388671875]], [[5.0]]]}, "model.layers.0.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.1884765625]]]}, "model.layers.0.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[404.0]]]}, "model.layers.0.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[123.0]]]}, "model.layers.0.self_attn.attn.impl.k_cache": {"inputs": [[[5.0]]]}, "model.layers.0.self_attn.attn.impl.v_cache": {"inputs": [[[0.1884765625]]]}, "model.layers.0.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.71875]], [[5.0]], [[0.1884765625]]], "outputs": [[[0.12353515625]], [[1.0]]]}, "model.layers.0.mlp.gate_up_proj": {"inputs": [[[1.453125]]], "params": {"weight": [[43.25]]}}, "model.layers.0.mlp.down_proj": {"inputs": [[[0.5390625]]], "outputs": [[[8.625]], [[52.5]]], "params": {"weight": [[0.64453125]]}}, "model.layers.1.self_attn.qkv_proj": {"inputs": [[[7.40625]]], "params": {"weight": [[4.875]]}}, "model.layers.1.self_attn.o_proj": {"inputs": [[[0.158203125]]], "outputs": [[[0.65625]], [[0.765625]]], "params": {"weight": [[0.515625]]}}, "model.layers.1.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.85546875]], [[12.4375]]]}, "model.layers.1.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.8359375]]]}, "model.layers.1.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[3.609375]]]}, "model.layers.1.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.984375]]]}, "model.layers.1.self_attn.attn.impl.k_cache": {"inputs": [[[12.4375]]]}, "model.layers.1.self_attn.attn.impl.v_cache": {"inputs": [[[0.8359375]]]}, "model.layers.1.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.9375]], [[12.4375]], [[0.8203125]]], "outputs": [[[0.1123046875]], [[1.0]]]}, "model.layers.1.mlp.gate_up_proj": {"inputs": [[[3.09375]]], "params": {"weight": [[7.875]]}}, "model.layers.1.mlp.down_proj": {"inputs": [[[6.5625]]], "outputs": [[[2.171875]], [[5.0625]]], "params": {"weight": [[0.52734375]]}}, "model.layers.2.self_attn.qkv_proj": {"inputs": [[[6.625]]], "params": {"weight": [[1.2890625]]}}, "model.layers.2.self_attn.o_proj": {"inputs": [[[0.06787109375]]], "outputs": [[[0.044189453125]], [[0.099609375]]], "params": {"weight": [[0.2158203125]]}}, "model.layers.2.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.0625]], [[12.1875]]]}, "model.layers.2.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.5234375]]]}, "model.layers.2.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[6.53125]]]}, "model.layers.2.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.65625]]]}, "model.layers.2.self_attn.attn.impl.k_cache": {"inputs": [[[12.1875]]]}, "model.layers.2.self_attn.attn.impl.v_cache": {"inputs": [[[0.5234375]]]}, "model.layers.2.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[11.4375]], [[11.75]], [[0.5078125]]], "outputs": [[[0.05078125]], [[1.0]]]}, "model.layers.2.mlp.gate_up_proj": {"inputs": [[[2.578125]]], "params": {"weight": [[2.78125]]}}, "model.layers.2.mlp.down_proj": {"inputs": [[[0.26171875]]], "outputs": [[[0.345703125]], [[0.72265625]]], "params": {"weight": [[0.298828125]]}}, "model.layers.3.self_attn.qkv_proj": {"inputs": [[[3.796875]]], "params": {"weight": [[7.875]]}}, "model.layers.3.self_attn.o_proj": {"inputs": [[[0.26953125]]], "outputs": [[[0.64453125]], [[0.9453125]]], "params": {"weight": [[0.26953125]]}}, "model.layers.3.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.48046875]], [[12.8125]]]}, "model.layers.3.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.53515625]]]}, "model.layers.3.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[5.375]]]}, "model.layers.3.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.140625]]]}, "model.layers.3.self_attn.attn.impl.k_cache": {"inputs": [[[12.8125]]]}, "model.layers.3.self_attn.attn.impl.v_cache": {"inputs": [[[0.53515625]]]}, "model.layers.3.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.0625]], [[12.0]], [[0.48046875]]], "outputs": [[[0.26953125]], [[1.0]]]}, "model.layers.3.mlp.gate_up_proj": {"inputs": [[[22.625]]], "params": {"weight": [[3.203125]]}}, "model.layers.3.mlp.down_proj": {"inputs": [[[1.0078125]]], "outputs": [[[4.90625]], [[452.0]]], "params": {"weight": [[1.3671875]]}}, "model.layers.4.self_attn.qkv_proj": {"inputs": [[[4.875]]], "params": {"weight": [[0.322265625]]}}, "model.layers.4.self_attn.o_proj": {"inputs": [[[0.16796875]]], "outputs": [[[0.1015625]], [[0.154296875]]], "params": {"weight": [[0.330078125]]}}, "model.layers.4.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.78515625]], [[13.75]]]}, "model.layers.4.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.875]]]}, "model.layers.4.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.375]]]}, "model.layers.4.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.8125]]]}, "model.layers.4.self_attn.attn.impl.k_cache": {"inputs": [[[13.75]]]}, "model.layers.4.self_attn.attn.impl.v_cache": {"inputs": [[[0.875]]]}, "model.layers.4.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.9375]], [[13.1875]], [[0.875]]], "outputs": [[[0.1328125]], [[1.0]]]}, "model.layers.4.mlp.gate_up_proj": {"inputs": [[[1.8828125]]], "params": {"weight": [[1.1328125]]}}, "model.layers.4.mlp.down_proj": {"inputs": [[[0.388671875]]], "outputs": [[[0.15234375]], [[0.68359375]]], "params": {"weight": [[1.0234375]]}}, "model.layers.5.self_attn.qkv_proj": {"inputs": [[[4.84375]]], "params": {"weight": [[3.1875]]}}, "model.layers.5.self_attn.o_proj": {"inputs": [[[0.146484375]]], "outputs": [[[0.06640625]], [[0.10888671875]]], "params": {"weight": [[0.09326171875]]}}, "model.layers.5.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.63671875]], [[18.875]]]}, "model.layers.5.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.76953125]]]}, "model.layers.5.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[3.5]]]}, "model.layers.5.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.015625]]]}, "model.layers.5.self_attn.attn.impl.k_cache": {"inputs": [[[18.875]]]}, "model.layers.5.self_attn.attn.impl.v_cache": {"inputs": [[[0.76953125]]]}, "model.layers.5.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.1875]], [[17.625]], [[0.76953125]]], "outputs": [[[0.107421875]], [[1.0]]]}, "model.layers.5.mlp.gate_up_proj": {"inputs": [[[3.890625]]], "params": {"weight": [[11.875]]}}, "model.layers.5.mlp.down_proj": {"inputs": [[[0.369140625]]], "outputs": [[[1.0078125]], [[7.375]]], "params": {"weight": [[0.66796875]]}}, "model.layers.6.self_attn.qkv_proj": {"inputs": [[[5.09375]]], "params": {"weight": [[0.345703125]]}}, "model.layers.6.self_attn.o_proj": {"inputs": [[[0.1064453125]]], "outputs": [[[0.2451171875]], [[0.279296875]]], "params": {"weight": [[0.146484375]]}}, "model.layers.6.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.515625]], [[15.75]]]}, "model.layers.6.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.94921875]]]}, "model.layers.6.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[3.3125]]]}, "model.layers.6.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.453125]]]}, "model.layers.6.self_attn.attn.impl.k_cache": {"inputs": [[[15.75]]]}, "model.layers.6.self_attn.attn.impl.v_cache": {"inputs": [[[0.94921875]]]}, "model.layers.6.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.21875]], [[14.8125]], [[0.86328125]]], "outputs": [[[0.0771484375]], [[1.0]]]}, "model.layers.6.mlp.gate_up_proj": {"inputs": [[[3.328125]]], "params": {"weight": [[1.015625]]}}, "model.layers.6.mlp.down_proj": {"inputs": [[[0.69140625]]], "outputs": [[[0.1982421875]], [[0.431640625]]], "params": {"weight": [[0.345703125]]}}, "model.layers.7.self_attn.qkv_proj": {"inputs": [[[5.625]]], "params": {"weight": [[0.357421875]]}}, "model.layers.7.self_attn.o_proj": {"inputs": [[[0.216796875]]], "outputs": [[[0.0849609375]], [[0.1240234375]]], "params": {"weight": [[0.181640625]]}}, "model.layers.7.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.7734375]], [[12.875]]]}, "model.layers.7.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.9765625]]]}, "model.layers.7.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.9375]]]}, "model.layers.7.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.1875]]]}, "model.layers.7.self_attn.attn.impl.k_cache": {"inputs": [[[12.875]]]}, "model.layers.7.self_attn.attn.impl.v_cache": {"inputs": [[[0.9765625]]]}, "model.layers.7.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.1875]], [[12.1875]], [[0.921875]]], "outputs": [[[0.134765625]], [[1.0]]]}, "model.layers.7.mlp.gate_up_proj": {"inputs": [[[3.3125]]], "params": {"weight": [[0.4453125]]}}, "model.layers.7.mlp.down_proj": {"inputs": [[[1.8984375]]], "outputs": [[[0.59765625]], [[1.59375]]], "params": {"weight": [[0.62109375]]}}, "model.layers.8.self_attn.qkv_proj": {"inputs": [[[4.8125]]], "params": {"weight": [[0.333984375]]}}, "model.layers.8.self_attn.o_proj": {"inputs": [[[0.1728515625]]], "outputs": [[[0.091796875]], [[0.138671875]]], "params": {"weight": [[0.12158203125]]}}, "model.layers.8.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.71875]], [[17.125]]]}, "model.layers.8.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.7109375]]]}, "model.layers.8.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[6.78125]]]}, "model.layers.8.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.890625]]]}, "model.layers.8.self_attn.attn.impl.k_cache": {"inputs": [[[17.125]]]}, "model.layers.8.self_attn.attn.impl.v_cache": {"inputs": [[[0.7109375]]]}, "model.layers.8.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.0]], [[16.5]], [[0.66015625]]], "outputs": [[[0.1728515625]], [[1.0]]]}, "model.layers.8.mlp.gate_up_proj": {"inputs": [[[3.078125]]], "params": {"weight": [[3.09375]]}}, "model.layers.8.mlp.down_proj": {"inputs": [[[0.51953125]]], "outputs": [[[0.126953125]], [[0.474609375]]], "params": {"weight": [[0.240234375]]}}, "model.layers.9.self_attn.qkv_proj": {"inputs": [[[5.59375]]], "params": {"weight": [[0.515625]]}}, "model.layers.9.self_attn.o_proj": {"inputs": [[[0.16015625]]], "outputs": [[[0.0517578125]], [[0.126953125]]], "params": {"weight": [[0.130859375]]}}, "model.layers.9.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6015625]], [[22.25]]]}, "model.layers.9.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.640625]]]}, "model.layers.9.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[5.96875]]]}, "model.layers.9.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.0]]]}, "model.layers.9.self_attn.attn.impl.k_cache": {"inputs": [[[22.25]]]}, "model.layers.9.self_attn.attn.impl.v_cache": {"inputs": [[[0.640625]]]}, "model.layers.9.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.0625]], [[21.125]], [[0.5234375]]], "outputs": [[[0.16015625]], [[1.0]]]}, "model.layers.9.mlp.gate_up_proj": {"inputs": [[[4.28125]]], "params": {"weight": [[15.4375]]}}, "model.layers.9.mlp.down_proj": {"inputs": [[[0.9140625]]], "outputs": [[[1.4375]], [[7.15625]]], "params": {"weight": [[0.2236328125]]}}, "model.layers.10.self_attn.qkv_proj": {"inputs": [[[6.4375]]], "params": {"weight": [[0.328125]]}}, "model.layers.10.self_attn.o_proj": {"inputs": [[[0.359375]]], "outputs": [[[0.11181640625]], [[0.234375]]], "params": {"weight": [[0.248046875]]}}, "model.layers.10.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6171875]], [[11.9375]]]}, "model.layers.10.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.234375]]]}, "model.layers.10.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[5.84375]]]}, "model.layers.10.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.765625]]]}, "model.layers.10.self_attn.attn.impl.k_cache": {"inputs": [[[11.9375]]]}, "model.layers.10.self_attn.attn.impl.v_cache": {"inputs": [[[1.234375]]]}, "model.layers.10.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.25]], [[11.5]], [[1.234375]]], "outputs": [[[0.1494140625]], [[1.0]]]}, "model.layers.10.mlp.gate_up_proj": {"inputs": [[[3.125]]], "params": {"weight": [[0.400390625]]}}, "model.layers.10.mlp.down_proj": {"inputs": [[[0.92578125]]], "outputs": [[[0.10888671875]], [[0.40625]]], "params": {"weight": [[0.31640625]]}}, "model.layers.11.self_attn.qkv_proj": {"inputs": [[[7.4375]]], "params": {"weight": [[2.8125]]}}, "model.layers.11.self_attn.o_proj": {"inputs": [[[0.20703125]]], "outputs": [[[0.08154296875]], [[0.166015625]]], "params": {"weight": [[0.193359375]]}}, "model.layers.11.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.640625]], [[15.5625]]]}, "model.layers.11.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.89453125]]]}, "model.layers.11.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[6.09375]]]}, "model.layers.11.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.84375]]]}, "model.layers.11.self_attn.attn.impl.k_cache": {"inputs": [[[15.5625]]]}, "model.layers.11.self_attn.attn.impl.v_cache": {"inputs": [[[0.89453125]]]}, "model.layers.11.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.0625]], [[15.125]], [[0.8203125]]], "outputs": [[[0.1611328125]], [[1.0]]]}, "model.layers.11.mlp.gate_up_proj": {"inputs": [[[3.34375]]], "params": {"weight": [[1.6015625]]}}, "model.layers.11.mlp.down_proj": {"inputs": [[[0.9921875]]], "outputs": [[[0.169921875]], [[0.75390625]]], "params": {"weight": [[0.2001953125]]}}, "model.layers.12.self_attn.qkv_proj": {"inputs": [[[6.1875]]], "params": {"weight": [[0.37109375]]}}, "model.layers.12.self_attn.o_proj": {"inputs": [[[0.357421875]]], "outputs": [[[0.1376953125]], [[0.32421875]]], "params": {"weight": [[0.1962890625]]}}, "model.layers.12.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.8515625]], [[13.25]]]}, "model.layers.12.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.796875]]]}, "model.layers.12.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.5]]]}, "model.layers.12.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.03125]]]}, "model.layers.12.self_attn.attn.impl.k_cache": {"inputs": [[[13.25]]]}, "model.layers.12.self_attn.attn.impl.v_cache": {"inputs": [[[0.796875]]]}, "model.layers.12.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.375]], [[12.625]], [[0.8046875]]], "outputs": [[[0.291015625]], [[1.0]]]}, "model.layers.12.mlp.gate_up_proj": {"inputs": [[[3.703125]]], "params": {"weight": [[0.8203125]]}}, "model.layers.12.mlp.down_proj": {"inputs": [[[0.72265625]]], "outputs": [[[0.5234375]], [[1.734375]]], "params": {"weight": [[0.326171875]]}}, "model.layers.13.self_attn.qkv_proj": {"inputs": [[[6.0]]], "params": {"weight": [[0.2578125]]}}, "model.layers.13.self_attn.o_proj": {"inputs": [[[0.3671875]]], "outputs": [[[0.150390625]], [[0.34375]]], "params": {"weight": [[0.16015625]]}}, "model.layers.13.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.62890625]], [[15.1875]]]}, "model.layers.13.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.859375]]]}, "model.layers.13.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[3.96875]]]}, "model.layers.13.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[2.859375]]]}, "model.layers.13.self_attn.attn.impl.k_cache": {"inputs": [[[15.1875]]]}, "model.layers.13.self_attn.attn.impl.v_cache": {"inputs": [[[1.859375]]]}, "model.layers.13.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.46875]], [[14.9375]], [[1.328125]]], "outputs": [[[0.333984375]], [[1.0]]]}, "model.layers.13.mlp.gate_up_proj": {"inputs": [[[3.8125]]], "params": {"weight": [[0.25390625]]}}, "model.layers.13.mlp.down_proj": {"inputs": [[[0.98828125]]], "outputs": [[[0.328125]], [[0.89453125]]], "params": {"weight": [[0.240234375]]}}, "model.layers.14.self_attn.qkv_proj": {"inputs": [[[6.4375]]], "params": {"weight": [[0.2197265625]]}}, "model.layers.14.self_attn.o_proj": {"inputs": [[[0.240234375]]], "outputs": [[[0.27734375]], [[0.490234375]]], "params": {"weight": [[0.1845703125]]}}, "model.layers.14.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.63671875]], [[10.125]]]}, "model.layers.14.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.6171875]]]}, "model.layers.14.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[4.25]]]}, "model.layers.14.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[3.375]]]}, "model.layers.14.self_attn.attn.impl.k_cache": {"inputs": [[[10.125]]]}, "model.layers.14.self_attn.attn.impl.v_cache": {"inputs": [[[1.6171875]]]}, "model.layers.14.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.5625]], [[9.8125]], [[1.4921875]]], "outputs": [[[0.2138671875]], [[1.0]]]}, "model.layers.14.mlp.gate_up_proj": {"inputs": [[[3.875]]], "params": {"weight": [[0.255859375]]}}, "model.layers.14.mlp.down_proj": {"inputs": [[[0.94140625]]], "outputs": [[[0.306640625]], [[0.87109375]]], "params": {"weight": [[0.330078125]]}}, "model.layers.15.self_attn.qkv_proj": {"inputs": [[[7.0]]], "params": {"weight": [[0.2255859375]]}}, "model.layers.15.self_attn.o_proj": {"inputs": [[[0.427734375]]], "outputs": [[[0.1884765625]], [[0.4140625]]], "params": {"weight": [[0.1357421875]]}}, "model.layers.15.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.7109375]], [[12.3125]]]}, "model.layers.15.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.5625]]]}, "model.layers.15.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[11.5625]]]}, "model.layers.15.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.75]]]}, "model.layers.15.self_attn.attn.impl.k_cache": {"inputs": [[[12.3125]]]}, "model.layers.15.self_attn.attn.impl.v_cache": {"inputs": [[[1.5625]]]}, "model.layers.15.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.71875]], [[11.25]], [[1.234375]]], "outputs": [[[0.306640625]], [[1.0]]]}, "model.layers.15.mlp.gate_up_proj": {"inputs": [[[3.453125]]], "params": {"weight": [[0.30078125]]}}, "model.layers.15.mlp.down_proj": {"inputs": [[[1.2734375]]], "outputs": [[[0.11328125]], [[0.375]]], "params": {"weight": [[0.62109375]]}}, "model.layers.16.self_attn.qkv_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.431640625]]}}, "model.layers.16.self_attn.o_proj": {"inputs": [[[0.53125]]], "outputs": [[[0.263671875]], [[0.5234375]]], "params": {"weight": [[0.1953125]]}}, "model.layers.16.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.625]], [[16.0]]]}, "model.layers.16.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.984375]]]}, "model.layers.16.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[10.0625]]]}, "model.layers.16.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.875]]]}, "model.layers.16.self_attn.attn.impl.k_cache": {"inputs": [[[16.0]]]}, "model.layers.16.self_attn.attn.impl.v_cache": {"inputs": [[[1.984375]]]}, "model.layers.16.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.3125]], [[15.375]], [[1.4921875]]], "outputs": [[[0.359375]], [[1.0]]]}, "model.layers.16.mlp.gate_up_proj": {"inputs": [[[4.1875]]], "params": {"weight": [[0.5078125]]}}, "model.layers.16.mlp.down_proj": {"inputs": [[[1.15625]]], "outputs": [[[0.32421875]], [[0.6328125]]], "params": {"weight": [[0.486328125]]}}, "model.layers.17.self_attn.qkv_proj": {"inputs": [[[10.9375]]], "params": {"weight": [[0.3515625]]}}, "model.layers.17.self_attn.o_proj": {"inputs": [[[0.71875]]], "outputs": [[[0.236328125]], [[0.75]]], "params": {"weight": [[0.1044921875]]}}, "model.layers.17.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.72265625]], [[13.0625]]]}, "model.layers.17.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.109375]]]}, "model.layers.17.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[37.25]]]}, "model.layers.17.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[17.625]]]}, "model.layers.17.self_attn.attn.impl.k_cache": {"inputs": [[[13.0625]]]}, "model.layers.17.self_attn.attn.impl.v_cache": {"inputs": [[[1.109375]]]}, "model.layers.17.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.6875]], [[12.0625]], [[1.0234375]]], "outputs": [[[0.43359375]], [[1.0]]]}, "model.layers.17.mlp.gate_up_proj": {"inputs": [[[4.03125]]], "params": {"weight": [[0.380859375]]}}, "model.layers.17.mlp.down_proj": {"inputs": [[[1.4765625]]], "outputs": [[[0.2353515625]], [[0.84765625]]], "params": {"weight": [[0.546875]]}}, "model.layers.18.self_attn.qkv_proj": {"inputs": [[[11.25]]], "params": {"weight": [[0.322265625]]}}, "model.layers.18.self_attn.o_proj": {"inputs": [[[0.498046875]]], "outputs": [[[0.20703125]], [[0.640625]]], "params": {"weight": [[0.1142578125]]}}, "model.layers.18.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.70703125]], [[14.625]]]}, "model.layers.18.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.90625]]]}, "model.layers.18.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[34.5]]]}, "model.layers.18.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.125]]]}, "model.layers.18.self_attn.attn.impl.k_cache": {"inputs": [[[14.625]]]}, "model.layers.18.self_attn.attn.impl.v_cache": {"inputs": [[[1.90625]]]}, "model.layers.18.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.71875]], [[13.75]], [[1.0]]], "outputs": [[[0.392578125]], [[1.0]]]}, "model.layers.18.mlp.gate_up_proj": {"inputs": [[[7.5625]]], "params": {"weight": [[0.26171875]]}}, "model.layers.18.mlp.down_proj": {"inputs": [[[27.0]]], "outputs": [[[6.625]], [[30.625]]], "params": {"weight": [[1.109375]]}}, "model.layers.19.self_attn.qkv_proj": {"inputs": [[[10.6875]]], "params": {"weight": [[0.255859375]]}}, "model.layers.19.self_attn.o_proj": {"inputs": [[[0.4921875]]], "outputs": [[[0.28515625]], [[0.76171875]]], "params": {"weight": [[0.16796875]]}}, "model.layers.19.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.703125]], [[12.375]]]}, "model.layers.19.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.6640625]]]}, "model.layers.19.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[37.5]]]}, "model.layers.19.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[20.25]]]}, "model.layers.19.self_attn.attn.impl.k_cache": {"inputs": [[[12.375]]]}, "model.layers.19.self_attn.attn.impl.v_cache": {"inputs": [[[1.6640625]]]}, "model.layers.19.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.84375]], [[11.375]], [[1.5390625]]], "outputs": [[[0.306640625]], [[1.0]]]}, "model.layers.19.mlp.gate_up_proj": {"inputs": [[[4.09375]]], "params": {"weight": [[0.44921875]]}}, "model.layers.19.mlp.down_proj": {"inputs": [[[1.953125]]], "outputs": [[[0.44921875]], [[1.2265625]]], "params": {"weight": [[0.44140625]]}}, "model.layers.20.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.390625]]}}, "model.layers.20.self_attn.o_proj": {"inputs": [[[0.625]]], "outputs": [[[0.2451171875]], [[1.1484375]]], "params": {"weight": [[0.2119140625]]}}, "model.layers.20.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.7109375]], [[18.125]]]}, "model.layers.20.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.46875]]]}, "model.layers.20.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[35.25]]]}, "model.layers.20.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[20.875]]]}, "model.layers.20.self_attn.attn.impl.k_cache": {"inputs": [[[18.125]]]}, "model.layers.20.self_attn.attn.impl.v_cache": {"inputs": [[[1.46875]]]}, "model.layers.20.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.03125]], [[16.375]], [[1.3203125]]], "outputs": [[[0.40234375]], [[1.0]]]}, "model.layers.20.mlp.gate_up_proj": {"inputs": [[[3.578125]]], "params": {"weight": [[0.314453125]]}}, "model.layers.20.mlp.down_proj": {"inputs": [[[2.015625]]], "outputs": [[[0.73046875]], [[4.6875]]], "params": {"weight": [[0.59375]]}}, "model.layers.21.self_attn.qkv_proj": {"inputs": [[[11.4375]]], "params": {"weight": [[0.3359375]]}}, "model.layers.21.self_attn.o_proj": {"inputs": [[[0.84765625]]], "outputs": [[[0.2890625]], [[1.3984375]]], "params": {"weight": [[0.1806640625]]}}, "model.layers.21.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.69140625]], [[14.875]]]}, "model.layers.21.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.296875]]]}, "model.layers.21.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[25.5]]]}, "model.layers.21.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.25]]]}, "model.layers.21.self_attn.attn.impl.k_cache": {"inputs": [[[14.875]]]}, "model.layers.21.self_attn.attn.impl.v_cache": {"inputs": [[[1.296875]]]}, "model.layers.21.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.96875]], [[13.625]], [[1.21875]]], "outputs": [[[0.5078125]], [[1.0]]]}, "model.layers.21.mlp.gate_up_proj": {"inputs": [[[3.578125]]], "params": {"weight": [[0.416015625]]}}, "model.layers.21.mlp.down_proj": {"inputs": [[[2.171875]]], "outputs": [[[0.68359375]], [[5.0]]], "params": {"weight": [[1.1640625]]}}, "model.layers.22.self_attn.qkv_proj": {"inputs": [[[11.0]]], "params": {"weight": [[0.296875]]}}, "model.layers.22.self_attn.o_proj": {"inputs": [[[0.6171875]]], "outputs": [[[0.37109375]], [[1.25]]], "params": {"weight": [[0.380859375]]}}, "model.layers.22.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.76953125]], [[15.1875]]]}, "model.layers.22.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.296875]]]}, "model.layers.22.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[20.875]]]}, "model.layers.22.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.625]]]}, "model.layers.22.self_attn.attn.impl.k_cache": {"inputs": [[[15.1875]]]}, "model.layers.22.self_attn.attn.impl.v_cache": {"inputs": [[[1.296875]]]}, "model.layers.22.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.0625]], [[14.1875]], [[1.0859375]]], "outputs": [[[0.4375]], [[1.0]]]}, "model.layers.22.mlp.gate_up_proj": {"inputs": [[[3.25]]], "params": {"weight": [[0.5859375]]}}, "model.layers.22.mlp.down_proj": {"inputs": [[[2.046875]]], "outputs": [[[0.478515625]], [[2.578125]]], "params": {"weight": [[0.482421875]]}}, "model.layers.23.self_attn.qkv_proj": {"inputs": [[[11.6875]]], "params": {"weight": [[0.33203125]]}}, "model.layers.23.self_attn.o_proj": {"inputs": [[[0.63671875]]], "outputs": [[[0.578125]], [[2.140625]]], "params": {"weight": [[0.380859375]]}}, "model.layers.23.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.78515625]], [[16.375]]]}, "model.layers.23.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.09375]]]}, "model.layers.23.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[31.25]]]}, "model.layers.23.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.0]]]}, "model.layers.23.self_attn.attn.impl.k_cache": {"inputs": [[[16.375]]]}, "model.layers.23.self_attn.attn.impl.v_cache": {"inputs": [[[1.09375]]]}, "model.layers.23.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.71875]], [[16.125]], [[1.0078125]]], "outputs": [[[0.51953125]], [[1.0]]]}, "model.layers.23.mlp.gate_up_proj": {"inputs": [[[3.265625]]], "params": {"weight": [[0.390625]]}}, "model.layers.23.mlp.down_proj": {"inputs": [[[6.375]]], "outputs": [[[0.5625]], [[2.75]]], "params": {"weight": [[0.396484375]]}}, "model.layers.24.self_attn.qkv_proj": {"inputs": [[[9.3125]]], "params": {"weight": [[0.375]]}}, "model.layers.24.self_attn.o_proj": {"inputs": [[[0.5703125]]], "outputs": [[[0.2158203125]], [[2.0]]], "params": {"weight": [[0.33984375]]}}, "model.layers.24.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6484375]], [[13.0]]]}, "model.layers.24.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.6875]]]}, "model.layers.24.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.1875]]]}, "model.layers.24.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.8125]]]}, "model.layers.24.self_attn.attn.impl.k_cache": {"inputs": [[[13.0]]]}, "model.layers.24.self_attn.attn.impl.v_cache": {"inputs": [[[1.6875]]]}, "model.layers.24.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.90625]], [[11.5625]], [[1.6875]]], "outputs": [[[0.546875]], [[1.0]]]}, "model.layers.24.mlp.gate_up_proj": {"inputs": [[[3.4375]]], "params": {"weight": [[0.50390625]]}}, "model.layers.24.mlp.down_proj": {"inputs": [[[2.421875]]], "outputs": [[[2.078125]], [[14.9375]]], "params": {"weight": [[0.79296875]]}}, "model.layers.25.self_attn.qkv_proj": {"inputs": [[[10.5625]]], "params": {"weight": [[0.33984375]]}}, "model.layers.25.self_attn.o_proj": {"inputs": [[[0.48828125]]], "outputs": [[[0.375]], [[2.28125]]], "params": {"weight": [[0.302734375]]}}, "model.layers.25.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.91015625]], [[14.4375]]]}, "model.layers.25.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.703125]]]}, "model.layers.25.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[35.5]]]}, "model.layers.25.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[21.375]]]}, "model.layers.25.self_attn.attn.impl.k_cache": {"inputs": [[[14.4375]]]}, "model.layers.25.self_attn.attn.impl.v_cache": {"inputs": [[[2.703125]]]}, "model.layers.25.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.0625]], [[14.4375]], [[2.703125]]], "outputs": [[[0.388671875]], [[1.0]]]}, "model.layers.25.mlp.gate_up_proj": {"inputs": [[[3.5]]], "params": {"weight": [[0.439453125]]}}, "model.layers.25.mlp.down_proj": {"inputs": [[[1.796875]]], "outputs": [[[0.9375]], [[6.28125]]], "params": {"weight": [[0.7734375]]}}, "model.layers.26.self_attn.qkv_proj": {"inputs": [[[10.875]]], "params": {"weight": [[0.353515625]]}}, "model.layers.26.self_attn.o_proj": {"inputs": [[[0.765625]]], "outputs": [[[0.341796875]], [[2.1875]]], "params": {"weight": [[0.400390625]]}}, "model.layers.26.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.78515625]], [[13.8125]]]}, "model.layers.26.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.015625]]]}, "model.layers.26.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[19.625]]]}, "model.layers.26.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.25]]]}, "model.layers.26.self_attn.attn.impl.k_cache": {"inputs": [[[13.8125]]]}, "model.layers.26.self_attn.attn.impl.v_cache": {"inputs": [[[2.015625]]]}, "model.layers.26.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.65625]], [[13.4375]], [[2.015625]]], "outputs": [[[0.44140625]], [[1.0]]]}, "model.layers.26.mlp.gate_up_proj": {"inputs": [[[3.890625]]], "params": {"weight": [[0.4296875]]}}, "model.layers.26.mlp.down_proj": {"inputs": [[[1.9921875]]], "outputs": [[[0.984375]], [[3.546875]]], "params": {"weight": [[0.7578125]]}}, "model.layers.27.self_attn.qkv_proj": {"inputs": [[[11.0]]], "params": {"weight": [[0.376953125]]}}, "model.layers.27.self_attn.o_proj": {"inputs": [[[0.53125]]], "outputs": [[[0.9921875]], [[2.34375]]], "params": {"weight": [[0.408203125]]}}, "model.layers.27.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.73828125]], [[15.8125]]]}, "model.layers.27.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.71875]]]}, "model.layers.27.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[23.125]]]}, "model.layers.27.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[12.5]]]}, "model.layers.27.self_attn.attn.impl.k_cache": {"inputs": [[[15.8125]]]}, "model.layers.27.self_attn.attn.impl.v_cache": {"inputs": [[[2.71875]]]}, "model.layers.27.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.90625]], [[13.5625]], [[2.71875]]], "outputs": [[[0.53125]], [[1.0]]]}, "model.layers.27.mlp.gate_up_proj": {"inputs": [[[4.0625]]], "params": {"weight": [[0.361328125]]}}, "model.layers.27.mlp.down_proj": {"inputs": [[[5.71875]]], "outputs": [[[0.85546875]], [[2.375]]], "params": {"weight": [[0.62890625]]}}, "model.layers.28.self_attn.qkv_proj": {"inputs": [[[11.0]]], "params": {"weight": [[0.3046875]]}}, "model.layers.28.self_attn.o_proj": {"inputs": [[[2.0]]], "outputs": [[[0.82421875]], [[2.625]]], "params": {"weight": [[0.875]]}}, "model.layers.28.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.73828125]], [[11.9375]]]}, "model.layers.28.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[6.59375]]]}, "model.layers.28.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[10.125]]]}, "model.layers.28.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.03125]]]}, "model.layers.28.self_attn.attn.impl.k_cache": {"inputs": [[[11.9375]]]}, "model.layers.28.self_attn.attn.impl.v_cache": {"inputs": [[[6.59375]]]}, "model.layers.28.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.65625]], [[10.0]], [[6.59375]]], "outputs": [[[2.0]], [[1.0]]]}, "model.layers.28.mlp.gate_up_proj": {"inputs": [[[4.78125]]], "params": {"weight": [[0.6328125]]}}, "model.layers.28.mlp.down_proj": {"inputs": [[[1.78125]]], "outputs": [[[0.8984375]], [[2.90625]]], "params": {"weight": [[0.92578125]]}}, "model.layers.29.self_attn.qkv_proj": {"inputs": [[[11.125]]], "params": {"weight": [[0.3828125]]}}, "model.layers.29.self_attn.o_proj": {"inputs": [[[0.78125]]], "outputs": [[[2.171875]], [[2.625]]], "params": {"weight": [[0.2451171875]]}}, "model.layers.29.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.890625]], [[15.375]]]}, "model.layers.29.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[5.4375]]]}, "model.layers.29.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[46.0]]]}, "model.layers.29.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[22.75]]]}, "model.layers.29.self_attn.attn.impl.k_cache": {"inputs": [[[15.375]]]}, "model.layers.29.self_attn.attn.impl.v_cache": {"inputs": [[[5.4375]]]}, "model.layers.29.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.3125]], [[15.375]], [[5.4375]]], "outputs": [[[0.69140625]], [[1.0]]]}, "model.layers.29.mlp.gate_up_proj": {"inputs": [[[4.9375]]], "params": {"weight": [[0.3359375]]}}, "model.layers.29.mlp.down_proj": {"inputs": [[[1.703125]]], "outputs": [[[1.046875]], [[2.390625]]], "params": {"weight": [[0.51171875]]}}, "model.layers.30.self_attn.qkv_proj": {"inputs": [[[10.25]]], "params": {"weight": [[0.5625]]}}, "model.layers.30.self_attn.o_proj": {"inputs": [[[0.5078125]]], "outputs": [[[0.38671875]], [[2.03125]]], "params": {"weight": [[0.31640625]]}}, "model.layers.30.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.87109375]], [[20.25]]]}, "model.layers.30.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.0859375]]]}, "model.layers.30.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[31.125]]]}, "model.layers.30.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.875]]]}, "model.layers.30.self_attn.attn.impl.k_cache": {"inputs": [[[20.25]]]}, "model.layers.30.self_attn.attn.impl.v_cache": {"inputs": [[[1.0859375]]]}, "model.layers.30.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.75]], [[18.375]], [[0.8046875]]], "outputs": [[[0.4921875]], [[1.0]]]}, "model.layers.30.mlp.gate_up_proj": {"inputs": [[[6.5]]], "params": {"weight": [[0.4921875]]}}, "model.layers.30.mlp.down_proj": {"inputs": [[[3.0]]], "outputs": [[[0.7890625]], [[3.375]]], "params": {"weight": [[0.6484375]]}}, "model.layers.31.self_attn.qkv_proj": {"inputs": [[[12.5]]], "params": {"weight": [[0.2314453125]]}}, "model.layers.31.self_attn.o_proj": {"inputs": [[[0.90625]]], "outputs": [[[1.4921875]], [[2.265625]]], "params": {"weight": [[0.400390625]]}}, "model.layers.31.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.828125]], [[17.625]]]}, "model.layers.31.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.9921875]]]}, "model.layers.31.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[37.5]]]}, "model.layers.31.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[19.375]]]}, "model.layers.31.self_attn.attn.impl.k_cache": {"inputs": [[[17.625]]]}, "model.layers.31.self_attn.attn.impl.v_cache": {"inputs": [[[1.9921875]]]}, "model.layers.31.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.53125]], [[17.125]], [[1.9921875]]], "outputs": [[[0.53125]], [[1.0]]]}, "model.layers.31.mlp.gate_up_proj": {"inputs": [[[5.09375]]], "params": {"weight": [[0.396484375]]}}, "model.layers.31.mlp.down_proj": {"inputs": [[[2.453125]]], "outputs": [[[0.60546875]], [[1.5390625]]], "params": {"weight": [[0.53515625]]}}, "model.layers.32.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.400390625]]}}, "model.layers.32.self_attn.o_proj": {"inputs": [[[1.0546875]]], "outputs": [[[0.470703125]], [[1.5]]], "params": {"weight": [[0.44140625]]}}, "model.layers.32.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6484375]], [[16.375]]]}, "model.layers.32.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.34375]]]}, "model.layers.32.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[35.75]]]}, "model.layers.32.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[19.25]]]}, "model.layers.32.self_attn.attn.impl.k_cache": {"inputs": [[[16.375]]]}, "model.layers.32.self_attn.attn.impl.v_cache": {"inputs": [[[4.34375]]]}, "model.layers.32.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.96875]], [[16.375]], [[4.34375]]], "outputs": [[[1.0546875]], [[1.0]]]}, "model.layers.32.mlp.gate_up_proj": {"inputs": [[[6.65625]]], "params": {"weight": [[0.3671875]]}}, "model.layers.32.mlp.down_proj": {"inputs": [[[4.28125]]], "outputs": [[[0.703125]], [[2.703125]]], "params": {"weight": [[1.015625]]}}, "model.layers.33.self_attn.qkv_proj": {"inputs": [[[14.25]]], "params": {"weight": [[0.421875]]}}, "model.layers.33.self_attn.o_proj": {"inputs": [[[0.75]]], "outputs": [[[1.4921875]], [[2.65625]]], "params": {"weight": [[0.490234375]]}}, "model.layers.33.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.7890625]], [[17.875]]]}, "model.layers.33.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.0625]]]}, "model.layers.33.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[25.0]]]}, "model.layers.33.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[17.25]]]}, "model.layers.33.self_attn.attn.impl.k_cache": {"inputs": [[[17.875]]]}, "model.layers.33.self_attn.attn.impl.v_cache": {"inputs": [[[2.0625]]]}, "model.layers.33.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.71875]], [[17.0]], [[2.0625]]], "outputs": [[[0.57421875]], [[1.0]]]}, "model.layers.33.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[28.125]]}}, "model.layers.33.mlp.down_proj": {"inputs": [[[3.203125]]], "outputs": [[[0.890625]], [[1.7578125]]], "params": {"weight": [[1.25]]}}, "model.layers.34.self_attn.qkv_proj": {"inputs": [[[13.25]]], "params": {"weight": [[0.337890625]]}}, "model.layers.34.self_attn.o_proj": {"inputs": [[[0.78125]]], "outputs": [[[13.5625]], [[14.9375]]], "params": {"weight": [[0.298828125]]}}, "model.layers.34.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.85546875]], [[18.125]]]}, "model.layers.34.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.71875]]]}, "model.layers.34.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[19.375]]]}, "model.layers.34.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.5]]]}, "model.layers.34.self_attn.attn.impl.k_cache": {"inputs": [[[18.125]]]}, "model.layers.34.self_attn.attn.impl.v_cache": {"inputs": [[[1.71875]]]}, "model.layers.34.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.75]], [[17.25]], [[1.71875]]], "outputs": [[[0.57421875]], [[1.0]]]}, "model.layers.34.mlp.gate_up_proj": {"inputs": [[[7.0]]], "params": {"weight": [[0.765625]]}}, "model.layers.34.mlp.down_proj": {"inputs": [[[1.9140625]]], "outputs": [[[0.85546875]], [[3.125]]], "params": {"weight": [[0.59375]]}}, "model.layers.35.self_attn.qkv_proj": {"inputs": [[[11.8125]]], "params": {"weight": [[0.271484375]]}}, "model.layers.35.self_attn.o_proj": {"inputs": [[[0.875]]], "outputs": [[[5.90625]], [[6.71875]]], "params": {"weight": [[0.96875]]}}, "model.layers.35.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.796875]], [[18.5]]]}, "model.layers.35.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.734375]]]}, "model.layers.35.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[35.75]]]}, "model.layers.35.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[17.875]]]}, "model.layers.35.self_attn.attn.impl.k_cache": {"inputs": [[[18.5]]]}, "model.layers.35.self_attn.attn.impl.v_cache": {"inputs": [[[2.734375]]]}, "model.layers.35.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.0]], [[16.875]], [[2.734375]]], "outputs": [[[0.6953125]], [[1.0]]]}, "model.layers.35.mlp.gate_up_proj": {"inputs": [[[6.5]]], "params": {"weight": [[0.388671875]]}}, "model.layers.35.mlp.down_proj": {"inputs": [[[3.34375]]], "outputs": [[[0.8125]], [[4.46875]]], "params": {"weight": [[0.328125]]}}, "model.layers.36.self_attn.qkv_proj": {"inputs": [[[15.4375]]], "params": {"weight": [[0.46875]]}}, "model.layers.36.self_attn.o_proj": {"inputs": [[[0.5234375]]], "outputs": [[[2.71875]], [[2.75]]], "params": {"weight": [[0.5390625]]}}, "model.layers.36.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.73046875]], [[17.375]]]}, "model.layers.36.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[5.90625]]]}, "model.layers.36.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[36.0]]]}, "model.layers.36.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.375]]]}, "model.layers.36.self_attn.attn.impl.k_cache": {"inputs": [[[17.375]]]}, "model.layers.36.self_attn.attn.impl.v_cache": {"inputs": [[[5.90625]]]}, "model.layers.36.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.25]], [[16.5]], [[5.90625]]], "outputs": [[[0.515625]], [[1.0]]]}, "model.layers.36.mlp.gate_up_proj": {"inputs": [[[8.125]]], "params": {"weight": [[0.458984375]]}}, "model.layers.36.mlp.down_proj": {"inputs": [[[2.671875]]], "outputs": [[[0.73828125]], [[3.40625]]], "params": {"weight": [[0.73828125]]}}, "model.layers.37.self_attn.qkv_proj": {"inputs": [[[17.625]]], "params": {"weight": [[0.291015625]]}}, "model.layers.37.self_attn.o_proj": {"inputs": [[[0.9921875]]], "outputs": [[[2.859375]], [[3.59375]]], "params": {"weight": [[1.0859375]]}}, "model.layers.37.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.73046875]], [[16.75]]]}, "model.layers.37.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[5.1875]]]}, "model.layers.37.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.0625]]]}, "model.layers.37.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.0625]]]}, "model.layers.37.self_attn.attn.impl.k_cache": {"inputs": [[[16.75]]]}, "model.layers.37.self_attn.attn.impl.v_cache": {"inputs": [[[5.1875]]]}, "model.layers.37.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.59375]], [[14.875]], [[5.1875]]], "outputs": [[[0.77734375]], [[1.0]]]}, "model.layers.37.mlp.gate_up_proj": {"inputs": [[[7.4375]]], "params": {"weight": [[0.42578125]]}}, "model.layers.37.mlp.down_proj": {"inputs": [[[2.671875]]], "outputs": [[[0.78515625]], [[2.21875]]], "params": {"weight": [[0.6171875]]}}, "model.layers.38.self_attn.qkv_proj": {"inputs": [[[16.125]]], "params": {"weight": [[0.97265625]]}}, "model.layers.38.self_attn.o_proj": {"inputs": [[[0.7421875]]], "outputs": [[[0.703125]], [[1.109375]]], "params": {"weight": [[0.34375]]}}, "model.layers.38.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.890625]], [[18.25]]]}, "model.layers.38.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.40625]]]}, "model.layers.38.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[44.0]]]}, "model.layers.38.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.4375]]]}, "model.layers.38.self_attn.attn.impl.k_cache": {"inputs": [[[18.25]]]}, "model.layers.38.self_attn.attn.impl.v_cache": {"inputs": [[[1.40625]]]}, "model.layers.38.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.125]], [[17.5]], [[1.0234375]]], "outputs": [[[0.7109375]], [[1.0]]]}, "model.layers.38.mlp.gate_up_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.376953125]]}}, "model.layers.38.mlp.down_proj": {"inputs": [[[7.09375]]], "outputs": [[[2.140625]], [[3.078125]]], "params": {"weight": [[0.9609375]]}}, "model.layers.39.self_attn.qkv_proj": {"inputs": [[[17.25]]], "params": {"weight": [[0.310546875]]}}, "model.layers.39.self_attn.o_proj": {"inputs": [[[0.90234375]]], "outputs": [[[3.515625]], [[3.625]]], "params": {"weight": [[0.40234375]]}}, "model.layers.39.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.60546875]], [[14.6875]]]}, "model.layers.39.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.640625]]]}, "model.layers.39.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[9.0]]]}, "model.layers.39.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.78125]]]}, "model.layers.39.self_attn.attn.impl.k_cache": {"inputs": [[[14.6875]]]}, "model.layers.39.self_attn.attn.impl.v_cache": {"inputs": [[[3.640625]]]}, "model.layers.39.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.84375]], [[13.75]], [[3.640625]]], "outputs": [[[0.70703125]], [[1.0]]]}, "model.layers.39.mlp.gate_up_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.5078125]]}}, "model.layers.39.mlp.down_proj": {"inputs": [[[2.359375]]], "outputs": [[[1.6328125]], [[3.203125]]], "params": {"weight": [[1.0859375]]}}, "model.layers.40.self_attn.qkv_proj": {"inputs": [[[13.0]]], "params": {"weight": [[0.32421875]]}}, "model.layers.40.self_attn.o_proj": {"inputs": [[[0.78515625]]], "outputs": [[[0.9609375]], [[2.0625]]], "params": {"weight": [[0.30078125]]}}, "model.layers.40.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.91796875]], [[10.5625]]]}, "model.layers.40.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.015625]]]}, "model.layers.40.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.875]]]}, "model.layers.40.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.8125]]]}, "model.layers.40.self_attn.attn.impl.k_cache": {"inputs": [[[10.5625]]]}, "model.layers.40.self_attn.attn.impl.v_cache": {"inputs": [[[2.015625]]]}, "model.layers.40.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.3125]], [[10.375]], [[1.140625]]], "outputs": [[[0.400390625]], [[1.0]]]}, "model.layers.40.mlp.gate_up_proj": {"inputs": [[[7.1875]]], "params": {"weight": [[0.404296875]]}}, "model.layers.40.mlp.down_proj": {"inputs": [[[3.40625]]], "outputs": [[[0.68359375]], [[1.34375]]], "params": {"weight": [[1.2734375]]}}, "model.layers.41.self_attn.qkv_proj": {"inputs": [[[16.25]]], "params": {"weight": [[0.384765625]]}}, "model.layers.41.self_attn.o_proj": {"inputs": [[[0.71875]]], "outputs": [[[0.2255859375]], [[1.078125]]], "params": {"weight": [[0.37890625]]}}, "model.layers.41.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.8828125]], [[15.0]]]}, "model.layers.41.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.4609375]]]}, "model.layers.41.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[29.75]]]}, "model.layers.41.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.375]]]}, "model.layers.41.self_attn.attn.impl.k_cache": {"inputs": [[[15.0]]]}, "model.layers.41.self_attn.attn.impl.v_cache": {"inputs": [[[1.4609375]]]}, "model.layers.41.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.6875]], [[14.0625]], [[1.0703125]]], "outputs": [[[0.63671875]], [[1.0]]]}, "model.layers.41.mlp.gate_up_proj": {"inputs": [[[7.28125]]], "params": {"weight": [[0.357421875]]}}, "model.layers.41.mlp.down_proj": {"inputs": [[[2.65625]]], "outputs": [[[0.9921875]], [[1.3984375]]], "params": {"weight": [[0.384765625]]}}, "model.layers.42.self_attn.qkv_proj": {"inputs": [[[14.8125]]], "params": {"weight": [[0.451171875]]}}, "model.layers.42.self_attn.o_proj": {"inputs": [[[0.2578125]]], "outputs": [[[0.1474609375]], [[0.5]]], "params": {"weight": [[0.2314453125]]}}, "model.layers.42.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.515625]], [[12.375]]]}, "model.layers.42.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.7890625]]]}, "model.layers.42.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[6.5625]]]}, "model.layers.42.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.84375]]]}, "model.layers.42.self_attn.attn.impl.k_cache": {"inputs": [[[12.375]]]}, "model.layers.42.self_attn.attn.impl.v_cache": {"inputs": [[[0.7890625]]]}, "model.layers.42.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.4375]], [[11.5]], [[0.8125]]], "outputs": [[[0.2578125]], [[1.0]]]}, "model.layers.42.mlp.gate_up_proj": {"inputs": [[[6.75]]], "params": {"weight": [[0.427734375]]}}, "model.layers.42.mlp.down_proj": {"inputs": [[[2.984375]]], "outputs": [[[0.85546875]], [[2.046875]]], "params": {"weight": [[0.267578125]]}}, "model.layers.43.self_attn.qkv_proj": {"inputs": [[[12.6875]]], "params": {"weight": [[0.388671875]]}}, "model.layers.43.self_attn.o_proj": {"inputs": [[[0.76171875]]], "outputs": [[[0.357421875]], [[0.734375]]], "params": {"weight": [[0.2333984375]]}}, "model.layers.43.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6875]], [[18.25]]]}, "model.layers.43.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.34375]]]}, "model.layers.43.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[17.75]]]}, "model.layers.43.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.6875]]]}, "model.layers.43.self_attn.attn.impl.k_cache": {"inputs": [[[18.25]]]}, "model.layers.43.self_attn.attn.impl.v_cache": {"inputs": [[[1.34375]]]}, "model.layers.43.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.4375]], [[15.0]], [[1.234375]]], "outputs": [[[0.546875]], [[1.0]]]}, "model.layers.43.mlp.gate_up_proj": {"inputs": [[[6.625]]], "params": {"weight": [[0.78125]]}}, "model.layers.43.mlp.down_proj": {"inputs": [[[3.03125]]], "outputs": [[[0.60546875]], [[0.81640625]]], "params": {"weight": [[0.384765625]]}}, "model.layers.44.self_attn.qkv_proj": {"inputs": [[[13.625]]], "params": {"weight": [[0.259765625]]}}, "model.layers.44.self_attn.o_proj": {"inputs": [[[0.94140625]]], "outputs": [[[0.67578125]], [[0.59375]]], "params": {"weight": [[0.21484375]]}}, "model.layers.44.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.91796875]], [[14.75]]]}, "model.layers.44.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.6796875]]]}, "model.layers.44.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[25.125]]]}, "model.layers.44.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[11.875]]]}, "model.layers.44.self_attn.attn.impl.k_cache": {"inputs": [[[14.75]]]}, "model.layers.44.self_attn.attn.impl.v_cache": {"inputs": [[[1.6796875]]]}, "model.layers.44.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.5]], [[14.1875]], [[1.1484375]]], "outputs": [[[0.76171875]], [[1.0]]]}, "model.layers.44.mlp.gate_up_proj": {"inputs": [[[7.09375]]], "params": {"weight": [[0.5703125]]}}, "model.layers.44.mlp.down_proj": {"inputs": [[[3.828125]]], "outputs": [[[0.486328125]], [[0.984375]]], "params": {"weight": [[0.92578125]]}}, "model.layers.45.self_attn.qkv_proj": {"inputs": [[[13.75]]], "params": {"weight": [[0.294921875]]}}, "model.layers.45.self_attn.o_proj": {"inputs": [[[0.6875]]], "outputs": [[[0.546875]], [[0.9921875]]], "params": {"weight": [[0.28125]]}}, "model.layers.45.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.56640625]], [[15.0]]]}, "model.layers.45.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.3828125]]]}, "model.layers.45.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.75]]]}, "model.layers.45.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.375]]]}, "model.layers.45.self_attn.attn.impl.k_cache": {"inputs": [[[15.0]]]}, "model.layers.45.self_attn.attn.impl.v_cache": {"inputs": [[[1.3828125]]]}, "model.layers.45.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.875]], [[13.9375]], [[0.90625]]], "outputs": [[[0.47265625]], [[1.0]]]}, "model.layers.45.mlp.gate_up_proj": {"inputs": [[[7.15625]]], "params": {"weight": [[0.46875]]}}, "model.layers.45.mlp.down_proj": {"inputs": [[[4.25]]], "outputs": [[[1.515625]], [[2.4375]]], "params": {"weight": [[1.5703125]]}}, "model.layers.46.self_attn.qkv_proj": {"inputs": [[[12.9375]]], "params": {"weight": [[0.71875]]}}, "model.layers.46.self_attn.o_proj": {"inputs": [[[0.359375]]], "outputs": [[[0.14453125]], [[0.2099609375]]], "params": {"weight": [[0.14453125]]}}, "model.layers.46.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.87109375]], [[16.75]]]}, "model.layers.46.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.5703125]]]}, "model.layers.46.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[9.75]]]}, "model.layers.46.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.625]]]}, "model.layers.46.self_attn.attn.impl.k_cache": {"inputs": [[[16.75]]]}, "model.layers.46.self_attn.attn.impl.v_cache": {"inputs": [[[0.5703125]]]}, "model.layers.46.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[10.125]], [[16.75]], [[0.49609375]]], "outputs": [[[0.33203125]], [[1.0]]]}, "model.layers.46.mlp.gate_up_proj": {"inputs": [[[6.4375]]], "params": {"weight": [[0.462890625]]}}, "model.layers.46.mlp.down_proj": {"inputs": [[[4.0625]]], "outputs": [[[0.359375]], [[0.671875]]], "params": {"weight": [[0.408203125]]}}, "model.layers.47.self_attn.qkv_proj": {"inputs": [[[11.9375]]], "params": {"weight": [[0.412109375]]}}, "model.layers.47.self_attn.o_proj": {"inputs": [[[0.3984375]]], "outputs": [[[0.625]], [[0.64453125]]], "params": {"weight": [[0.2333984375]]}}, "model.layers.47.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.55859375]], [[14.8125]]]}, "model.layers.47.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.0625]]]}, "model.layers.47.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[10.1875]]]}, "model.layers.47.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.34375]]]}, "model.layers.47.self_attn.attn.impl.k_cache": {"inputs": [[[14.8125]]]}, "model.layers.47.self_attn.attn.impl.v_cache": {"inputs": [[[2.0625]]]}, "model.layers.47.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.96875]], [[13.625]], [[2.0625]]], "outputs": [[[0.369140625]], [[1.0]]]}, "model.layers.47.mlp.gate_up_proj": {"inputs": [[[6.5625]]], "params": {"weight": [[0.392578125]]}}, "model.layers.47.mlp.down_proj": {"inputs": [[[4.28125]]], "outputs": [[[0.5234375]], [[1.21875]]], "params": {"weight": [[1.4375]]}}, "model.layers.48.self_attn.qkv_proj": {"inputs": [[[11.125]]], "params": {"weight": [[0.52734375]]}}, "model.layers.48.self_attn.o_proj": {"inputs": [[[0.60546875]]], "outputs": [[[0.3984375]], [[0.515625]]], "params": {"weight": [[0.259765625]]}}, "model.layers.48.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.70703125]], [[12.625]]]}, "model.layers.48.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.3515625]]]}, "model.layers.48.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[30.625]]]}, "model.layers.48.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.6875]]]}, "model.layers.48.self_attn.attn.impl.k_cache": {"inputs": [[[12.625]]]}, "model.layers.48.self_attn.attn.impl.v_cache": {"inputs": [[[1.3515625]]]}, "model.layers.48.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.5625]], [[10.0]], [[1.1015625]]], "outputs": [[[0.5703125]], [[1.0]]]}, "model.layers.48.mlp.gate_up_proj": {"inputs": [[[6.84375]]], "params": {"weight": [[0.62890625]]}}, "model.layers.48.mlp.down_proj": {"inputs": [[[4.125]]], "outputs": [[[0.447265625]], [[2.171875]]], "params": {"weight": [[1.2265625]]}}, "model.layers.49.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.37890625]]}}, "model.layers.49.self_attn.o_proj": {"inputs": [[[0.83984375]]], "outputs": [[[0.333984375]], [[0.4140625]]], "params": {"weight": [[0.51953125]]}}, "model.layers.49.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.63671875]], [[15.3125]]]}, "model.layers.49.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[6.53125]]]}, "model.layers.49.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[16.75]]]}, "model.layers.49.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.5625]]]}, "model.layers.49.self_attn.attn.impl.k_cache": {"inputs": [[[15.3125]]]}, "model.layers.49.self_attn.attn.impl.v_cache": {"inputs": [[[6.53125]]]}, "model.layers.49.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.59375]], [[12.9375]], [[6.53125]]], "outputs": [[[0.47265625]], [[1.0]]]}, "model.layers.49.mlp.gate_up_proj": {"inputs": [[[6.6875]]], "params": {"weight": [[0.60546875]]}}, "model.layers.49.mlp.down_proj": {"inputs": [[[4.40625]]], "outputs": [[[1.15625]], [[3.046875]]], "params": {"weight": [[0.359375]]}}, "model.layers.50.self_attn.qkv_proj": {"inputs": [[[11.125]]], "params": {"weight": [[0.4375]]}}, "model.layers.50.self_attn.o_proj": {"inputs": [[[0.45703125]]], "outputs": [[[0.2080078125]], [[0.4453125]]], "params": {"weight": [[0.1923828125]]}}, "model.layers.50.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5]], [[12.5]]]}, "model.layers.50.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.2578125]]]}, "model.layers.50.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.9375]]]}, "model.layers.50.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[8.0625]]]}, "model.layers.50.self_attn.attn.impl.k_cache": {"inputs": [[[12.5]]]}, "model.layers.50.self_attn.attn.impl.v_cache": {"inputs": [[[1.2578125]]]}, "model.layers.50.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.71875]], [[11.125]], [[1.2578125]]], "outputs": [[[0.45703125]], [[1.0]]]}, "model.layers.50.mlp.gate_up_proj": {"inputs": [[[6.53125]]], "params": {"weight": [[0.46484375]]}}, "model.layers.50.mlp.down_proj": {"inputs": [[[3.6875]]], "outputs": [[[0.92578125]], [[0.8125]]], "params": {"weight": [[0.37109375]]}}, "model.layers.51.self_attn.qkv_proj": {"inputs": [[[10.875]]], "params": {"weight": [[0.53125]]}}, "model.layers.51.self_attn.o_proj": {"inputs": [[[0.40625]]], "outputs": [[[0.396484375]], [[0.578125]]], "params": {"weight": [[0.1630859375]]}}, "model.layers.51.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5390625]], [[15.0]]]}, "model.layers.51.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.8984375]]]}, "model.layers.51.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[13.0]]]}, "model.layers.51.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.53125]]]}, "model.layers.51.self_attn.attn.impl.k_cache": {"inputs": [[[15.0]]]}, "model.layers.51.self_attn.attn.impl.v_cache": {"inputs": [[[0.8984375]]]}, "model.layers.51.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.03125]], [[14.625]], [[0.83203125]]], "outputs": [[[0.40625]], [[1.0]]]}, "model.layers.51.mlp.gate_up_proj": {"inputs": [[[6.46875]]], "params": {"weight": [[0.515625]]}}, "model.layers.51.mlp.down_proj": {"inputs": [[[4.5625]]], "outputs": [[[0.44921875]], [[2.796875]]], "params": {"weight": [[0.92578125]]}}, "model.layers.52.self_attn.qkv_proj": {"inputs": [[[12.375]]], "params": {"weight": [[0.37109375]]}}, "model.layers.52.self_attn.o_proj": {"inputs": [[[1.3203125]]], "outputs": [[[0.671875]], [[0.9375]]], "params": {"weight": [[0.3359375]]}}, "model.layers.52.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.984375]], [[12.375]]]}, "model.layers.52.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.5625]]]}, "model.layers.52.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[21.875]]]}, "model.layers.52.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.375]]]}, "model.layers.52.self_attn.attn.impl.k_cache": {"inputs": [[[12.375]]]}, "model.layers.52.self_attn.attn.impl.v_cache": {"inputs": [[[1.5625]]]}, "model.layers.52.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[10.3125]], [[12.25]], [[1.5625]]], "outputs": [[[0.69921875]], [[1.0]]]}, "model.layers.52.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[0.59375]]}}, "model.layers.52.mlp.down_proj": {"inputs": [[[5.125]]], "outputs": [[[0.85546875]], [[1.9453125]]], "params": {"weight": [[0.419921875]]}}, "model.layers.53.self_attn.qkv_proj": {"inputs": [[[11.25]]], "params": {"weight": [[0.498046875]]}}, "model.layers.53.self_attn.o_proj": {"inputs": [[[0.423828125]]], "outputs": [[[0.279296875]], [[0.63671875]]], "params": {"weight": [[0.3359375]]}}, "model.layers.53.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6171875]], [[14.0]]]}, "model.layers.53.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.859375]]]}, "model.layers.53.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[11.75]]]}, "model.layers.53.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.8125]]]}, "model.layers.53.self_attn.attn.impl.k_cache": {"inputs": [[[14.0]]]}, "model.layers.53.self_attn.attn.impl.v_cache": {"inputs": [[[3.859375]]]}, "model.layers.53.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.9375]], [[14.0]], [[3.859375]]], "outputs": [[[0.423828125]], [[1.0]]]}, "model.layers.53.mlp.gate_up_proj": {"inputs": [[[6.6875]]], "params": {"weight": [[0.5]]}}, "model.layers.53.mlp.down_proj": {"inputs": [[[4.75]]], "outputs": [[[0.59765625]], [[1.921875]]], "params": {"weight": [[0.5859375]]}}, "model.layers.54.self_attn.qkv_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.53515625]]}}, "model.layers.54.self_attn.o_proj": {"inputs": [[[0.640625]]], "outputs": [[[0.66796875]], [[1.078125]]], "params": {"weight": [[0.55859375]]}}, "model.layers.54.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.47265625]], [[13.125]]]}, "model.layers.54.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.484375]]]}, "model.layers.54.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[17.625]]]}, "model.layers.54.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[13.75]]]}, "model.layers.54.self_attn.attn.impl.k_cache": {"inputs": [[[13.125]]]}, "model.layers.54.self_attn.attn.impl.v_cache": {"inputs": [[[3.484375]]]}, "model.layers.54.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.875]], [[13.125]], [[3.484375]]], "outputs": [[[0.60546875]], [[1.0]]]}, "model.layers.54.mlp.gate_up_proj": {"inputs": [[[6.5625]]], "params": {"weight": [[0.490234375]]}}, "model.layers.54.mlp.down_proj": {"inputs": [[[9.6875]]], "outputs": [[[0.6171875]], [[1.2734375]]], "params": {"weight": [[0.22265625]]}}, "model.layers.55.self_attn.qkv_proj": {"inputs": [[[10.3125]]], "params": {"weight": [[0.61328125]]}}, "model.layers.55.self_attn.o_proj": {"inputs": [[[0.50390625]]], "outputs": [[[0.9765625]], [[1.28125]]], "params": {"weight": [[1.578125]]}}, "model.layers.55.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.48828125]], [[13.125]]]}, "model.layers.55.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.546875]]]}, "model.layers.55.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[13.4375]]]}, "model.layers.55.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.0]]]}, "model.layers.55.self_attn.attn.impl.k_cache": {"inputs": [[[13.125]]]}, "model.layers.55.self_attn.attn.impl.v_cache": {"inputs": [[[1.546875]]]}, "model.layers.55.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.5]], [[12.75]], [[1.546875]]], "outputs": [[[0.50390625]], [[1.0]]]}, "model.layers.55.mlp.gate_up_proj": {"inputs": [[[6.6875]]], "params": {"weight": [[0.458984375]]}}, "model.layers.55.mlp.down_proj": {"inputs": [[[6.0]]], "outputs": [[[1.59375]], [[3.078125]]], "params": {"weight": [[0.52734375]]}}, "model.layers.56.self_attn.qkv_proj": {"inputs": [[[13.6875]]], "params": {"weight": [[0.38671875]]}}, "model.layers.56.self_attn.o_proj": {"inputs": [[[0.57421875]]], "outputs": [[[0.72265625]], [[1.0546875]]], "params": {"weight": [[0.2080078125]]}}, "model.layers.56.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.55078125]], [[14.4375]]]}, "model.layers.56.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.078125]]]}, "model.layers.56.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[14.125]]]}, "model.layers.56.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.8125]]]}, "model.layers.56.self_attn.attn.impl.k_cache": {"inputs": [[[14.4375]]]}, "model.layers.56.self_attn.attn.impl.v_cache": {"inputs": [[[2.078125]]]}, "model.layers.56.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.78125]], [[14.4375]], [[2.078125]]], "outputs": [[[0.546875]], [[1.0]]]}, "model.layers.56.mlp.gate_up_proj": {"inputs": [[[6.90625]]], "params": {"weight": [[0.58984375]]}}, "model.layers.56.mlp.down_proj": {"inputs": [[[4.84375]]], "outputs": [[[2.109375]], [[2.296875]]], "params": {"weight": [[1.46875]]}}, "model.layers.57.self_attn.qkv_proj": {"inputs": [[[12.1875]]], "params": {"weight": [[0.396484375]]}}, "model.layers.57.self_attn.o_proj": {"inputs": [[[0.7890625]]], "outputs": [[[0.306640625]], [[0.7578125]]], "params": {"weight": [[0.185546875]]}}, "model.layers.57.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.6640625]], [[13.625]]]}, "model.layers.57.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.0625]]]}, "model.layers.57.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[15.6875]]]}, "model.layers.57.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.3125]]]}, "model.layers.57.self_attn.attn.impl.k_cache": {"inputs": [[[13.625]]]}, "model.layers.57.self_attn.attn.impl.v_cache": {"inputs": [[[1.0625]]]}, "model.layers.57.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.5]], [[12.625]], [[1.0625]]], "outputs": [[[0.69140625]], [[1.0]]]}, "model.layers.57.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[0.384765625]]}}, "model.layers.57.mlp.down_proj": {"inputs": [[[4.3125]]], "outputs": [[[0.66015625]], [[1.9609375]]], "params": {"weight": [[0.275390625]]}}, "model.layers.58.self_attn.qkv_proj": {"inputs": [[[12.1875]]], "params": {"weight": [[0.578125]]}}, "model.layers.58.self_attn.o_proj": {"inputs": [[[0.482421875]]], "outputs": [[[0.298828125]], [[0.66796875]]], "params": {"weight": [[0.2041015625]]}}, "model.layers.58.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.466796875]], [[13.0625]]]}, "model.layers.58.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.80859375]]]}, "model.layers.58.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[8.75]]]}, "model.layers.58.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.4375]]]}, "model.layers.58.self_attn.attn.impl.k_cache": {"inputs": [[[13.0625]]]}, "model.layers.58.self_attn.attn.impl.v_cache": {"inputs": [[[0.80859375]]]}, "model.layers.58.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.65625]], [[12.75]], [[0.67578125]]], "outputs": [[[0.326171875]], [[1.0]]]}, "model.layers.58.mlp.gate_up_proj": {"inputs": [[[6.875]]], "params": {"weight": [[0.4140625]]}}, "model.layers.58.mlp.down_proj": {"inputs": [[[9.625]]], "outputs": [[[0.423828125]], [[0.734375]]], "params": {"weight": [[0.392578125]]}}, "model.layers.59.self_attn.qkv_proj": {"inputs": [[[10.8125]]], "params": {"weight": [[0.58203125]]}}, "model.layers.59.self_attn.o_proj": {"inputs": [[[0.73828125]]], "outputs": [[[0.251953125]], [[1.0625]]], "params": {"weight": [[0.1611328125]]}}, "model.layers.59.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.47265625]], [[17.75]]]}, "model.layers.59.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.3359375]]]}, "model.layers.59.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[18.75]]]}, "model.layers.59.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.375]]]}, "model.layers.59.self_attn.attn.impl.k_cache": {"inputs": [[[17.75]]]}, "model.layers.59.self_attn.attn.impl.v_cache": {"inputs": [[[1.3359375]]]}, "model.layers.59.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.5625]], [[13.375]], [[1.2734375]]], "outputs": [[[0.73828125]], [[1.0]]]}, "model.layers.59.mlp.gate_up_proj": {"inputs": [[[7.21875]]], "params": {"weight": [[0.53125]]}}, "model.layers.59.mlp.down_proj": {"inputs": [[[5.78125]]], "outputs": [[[0.578125]], [[7.1875]]], "params": {"weight": [[1.6484375]]}}, "model.layers.60.self_attn.qkv_proj": {"inputs": [[[13.3125]]], "params": {"weight": [[0.349609375]]}}, "model.layers.60.self_attn.o_proj": {"inputs": [[[1.09375]]], "outputs": [[[0.484375]], [[0.58984375]]], "params": {"weight": [[0.1298828125]]}}, "model.layers.60.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.0390625]], [[15.75]]]}, "model.layers.60.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.9140625]]]}, "model.layers.60.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[16.25]]]}, "model.layers.60.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[9.1875]]]}, "model.layers.60.self_attn.attn.impl.k_cache": {"inputs": [[[15.75]]]}, "model.layers.60.self_attn.attn.impl.v_cache": {"inputs": [[[1.9140625]]]}, "model.layers.60.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[10.25]], [[15.75]], [[1.8359375]]], "outputs": [[[0.6953125]], [[1.0]]]}, "model.layers.60.mlp.gate_up_proj": {"inputs": [[[7.34375]]], "params": {"weight": [[0.380859375]]}}, "model.layers.60.mlp.down_proj": {"inputs": [[[7.375]]], "outputs": [[[2.296875]], [[3.671875]]], "params": {"weight": [[0.498046875]]}}, "model.layers.61.self_attn.qkv_proj": {"inputs": [[[11.5]]], "params": {"weight": [[0.59375]]}}, "model.layers.61.self_attn.o_proj": {"inputs": [[[0.703125]]], "outputs": [[[0.26953125]], [[0.62109375]]], "params": {"weight": [[0.07763671875]]}}, "model.layers.61.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.53515625]], [[14.125]]]}, "model.layers.61.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.1328125]]]}, "model.layers.61.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[12.125]]]}, "model.layers.61.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.09375]]]}, "model.layers.61.self_attn.attn.impl.k_cache": {"inputs": [[[14.125]]]}, "model.layers.61.self_attn.attn.impl.v_cache": {"inputs": [[[1.1328125]]]}, "model.layers.61.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.625]], [[12.9375]], [[0.890625]]], "outputs": [[[0.61328125]], [[1.0]]]}, "model.layers.61.mlp.gate_up_proj": {"inputs": [[[7.46875]]], "params": {"weight": [[0.373046875]]}}, "model.layers.61.mlp.down_proj": {"inputs": [[[6.5]]], "outputs": [[[0.7421875]], [[2.421875]]], "params": {"weight": [[0.474609375]]}}, "model.layers.62.self_attn.qkv_proj": {"inputs": [[[11.5625]]], "params": {"weight": [[0.76953125]]}}, "model.layers.62.self_attn.o_proj": {"inputs": [[[0.291015625]]], "outputs": [[[0.75]], [[0.76171875]]], "params": {"weight": [[0.0927734375]]}}, "model.layers.62.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.76171875]], [[16.75]]]}, "model.layers.62.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[0.5234375]]]}, "model.layers.62.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[7.53125]]]}, "model.layers.62.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.375]]]}, "model.layers.62.self_attn.attn.impl.k_cache": {"inputs": [[[16.75]]]}, "model.layers.62.self_attn.attn.impl.v_cache": {"inputs": [[[0.5234375]]]}, "model.layers.62.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.1875]], [[16.75]], [[0.5625]]], "outputs": [[[0.291015625]], [[1.0]]]}, "model.layers.62.mlp.gate_up_proj": {"inputs": [[[7.625]]], "params": {"weight": [[0.5]]}}, "model.layers.62.mlp.down_proj": {"inputs": [[[9.0]]], "outputs": [[[1.0078125]], [[1.3359375]]], "params": {"weight": [[1.1953125]]}}, "model.layers.63.self_attn.qkv_proj": {"inputs": [[[10.1875]]], "params": {"weight": [[0.8125]]}}, "model.layers.63.self_attn.o_proj": {"inputs": [[[0.9296875]]], "outputs": [[[0.259765625]], [[1.1875]]], "params": {"weight": [[0.25390625]]}}, "model.layers.63.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.5703125]], [[12.0]]]}, "model.layers.63.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.421875]]]}, "model.layers.63.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[27.625]]]}, "model.layers.63.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.8125]]]}, "model.layers.63.self_attn.attn.impl.k_cache": {"inputs": [[[12.0]]]}, "model.layers.63.self_attn.attn.impl.v_cache": {"inputs": [[[1.421875]]]}, "model.layers.63.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.5]], [[10.3125]], [[1.3203125]]], "outputs": [[[0.69140625]], [[1.0]]]}, "model.layers.63.mlp.gate_up_proj": {"inputs": [[[7.9375]]], "params": {"weight": [[0.58203125]]}}, "model.layers.63.mlp.down_proj": {"inputs": [[[6.9375]]], "outputs": [[[0.73828125]], [[1.640625]]], "params": {"weight": [[0.28125]]}}, "model.layers.64.self_attn.qkv_proj": {"inputs": [[[13.0]]], "params": {"weight": [[0.337890625]]}}, "model.layers.64.self_attn.o_proj": {"inputs": [[[1.1875]]], "outputs": [[[0.341796875]], [[1.5]]], "params": {"weight": [[0.14453125]]}}, "model.layers.64.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.1953125]], [[12.9375]]]}, "model.layers.64.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.453125]]]}, "model.layers.64.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[5.78125]]]}, "model.layers.64.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[4.9375]]]}, "model.layers.64.self_attn.attn.impl.k_cache": {"inputs": [[[12.9375]]]}, "model.layers.64.self_attn.attn.impl.v_cache": {"inputs": [[[2.453125]]]}, "model.layers.64.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[12.25]], [[12.75]], [[2.09375]]], "outputs": [[[1.1171875]], [[1.0]]]}, "model.layers.64.mlp.gate_up_proj": {"inputs": [[[8.3125]]], "params": {"weight": [[0.423828125]]}}, "model.layers.64.mlp.down_proj": {"inputs": [[[9.1875]]], "outputs": [[[0.578125]], [[2.578125]]], "params": {"weight": [[0.98046875]]}}, "model.layers.65.self_attn.qkv_proj": {"inputs": [[[14.6875]]], "params": {"weight": [[0.81640625]]}}, "model.layers.65.self_attn.o_proj": {"inputs": [[[0.5234375]]], "outputs": [[[1.75]], [[1.6875]]], "params": {"weight": [[0.42578125]]}}, "model.layers.65.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.494140625]], [[20.125]]]}, "model.layers.65.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.9375]]]}, "model.layers.65.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[10.5625]]]}, "model.layers.65.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.75]]]}, "model.layers.65.self_attn.attn.impl.k_cache": {"inputs": [[[20.125]]]}, "model.layers.65.self_attn.attn.impl.v_cache": {"inputs": [[[4.9375]]]}, "model.layers.65.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[5.28125]], [[20.125]], [[4.9375]]], "outputs": [[[0.5234375]], [[1.0]]]}, "model.layers.65.mlp.gate_up_proj": {"inputs": [[[8.125]]], "params": {"weight": [[0.375]]}}, "model.layers.65.mlp.down_proj": {"inputs": [[[14.9375]]], "outputs": [[[0.76171875]], [[2.015625]]], "params": {"weight": [[0.6640625]]}}, "model.layers.66.self_attn.qkv_proj": {"inputs": [[[11.375]]], "params": {"weight": [[0.71484375]]}}, "model.layers.66.self_attn.o_proj": {"inputs": [[[0.59375]]], "outputs": [[[1.109375]], [[1.3359375]]], "params": {"weight": [[0.208984375]]}}, "model.layers.66.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.44140625]], [[14.375]]]}, "model.layers.66.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.46875]]]}, "model.layers.66.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[9.0625]]]}, "model.layers.66.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.75]]]}, "model.layers.66.self_attn.attn.impl.k_cache": {"inputs": [[[14.375]]]}, "model.layers.66.self_attn.attn.impl.v_cache": {"inputs": [[[2.46875]]]}, "model.layers.66.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[4.71875]], [[14.375]], [[2.46875]]], "outputs": [[[0.4921875]], [[1.0]]]}, "model.layers.66.mlp.gate_up_proj": {"inputs": [[[7.78125]]], "params": {"weight": [[0.49609375]]}}, "model.layers.66.mlp.down_proj": {"inputs": [[[7.125]]], "outputs": [[[2.5625]], [[3.5]]], "params": {"weight": [[0.50390625]]}}, "model.layers.67.self_attn.qkv_proj": {"inputs": [[[12.6875]]], "params": {"weight": [[0.427734375]]}}, "model.layers.67.self_attn.o_proj": {"inputs": [[[1.3515625]]], "outputs": [[[0.62890625]], [[1.453125]]], "params": {"weight": [[0.1962890625]]}}, "model.layers.67.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.921875]], [[17.75]]]}, "model.layers.67.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.09375]]]}, "model.layers.67.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[22.75]]]}, "model.layers.67.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[15.25]]]}, "model.layers.67.self_attn.attn.impl.k_cache": {"inputs": [[[17.75]]]}, "model.layers.67.self_attn.attn.impl.v_cache": {"inputs": [[[2.09375]]]}, "model.layers.67.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.1875]], [[17.75]], [[1.9453125]]], "outputs": [[[1.015625]], [[1.0]]]}, "model.layers.67.mlp.gate_up_proj": {"inputs": [[[8.1875]]], "params": {"weight": [[0.390625]]}}, "model.layers.67.mlp.down_proj": {"inputs": [[[8.375]]], "outputs": [[[0.68359375]], [[1.5546875]]], "params": {"weight": [[0.427734375]]}}, "model.layers.68.self_attn.qkv_proj": {"inputs": [[[13.75]]], "params": {"weight": [[0.384765625]]}}, "model.layers.68.self_attn.o_proj": {"inputs": [[[1.046875]]], "outputs": [[[0.294921875]], [[2.09375]]], "params": {"weight": [[0.49609375]]}}, "model.layers.68.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.8671875]], [[15.0625]]]}, "model.layers.68.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.46875]]]}, "model.layers.68.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[10.5]]]}, "model.layers.68.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.125]]]}, "model.layers.68.self_attn.attn.impl.k_cache": {"inputs": [[[15.0625]]]}, "model.layers.68.self_attn.attn.impl.v_cache": {"inputs": [[[2.46875]]]}, "model.layers.68.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[8.375]], [[14.0625]], [[2.21875]]], "outputs": [[[1.015625]], [[1.0]]]}, "model.layers.68.mlp.gate_up_proj": {"inputs": [[[9.0625]]], "params": {"weight": [[0.369140625]]}}, "model.layers.68.mlp.down_proj": {"inputs": [[[11.75]]], "outputs": [[[1.6171875]], [[3.953125]]], "params": {"weight": [[1.34375]]}}, "model.layers.69.self_attn.qkv_proj": {"inputs": [[[12.0]]], "params": {"weight": [[0.2578125]]}}, "model.layers.69.self_attn.o_proj": {"inputs": [[[1.390625]]], "outputs": [[[0.58203125]], [[2.109375]]], "params": {"weight": [[0.1435546875]]}}, "model.layers.69.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.59765625]], [[13.4375]]]}, "model.layers.69.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.828125]]]}, "model.layers.69.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[6.125]]]}, "model.layers.69.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[5.28125]]]}, "model.layers.69.self_attn.attn.impl.k_cache": {"inputs": [[[13.4375]]]}, "model.layers.69.self_attn.attn.impl.v_cache": {"inputs": [[[1.828125]]]}, "model.layers.69.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[6.0625]], [[13.4375]], [[1.78125]]], "outputs": [[[0.921875]], [[1.0]]]}, "model.layers.69.mlp.gate_up_proj": {"inputs": [[[8.625]]], "params": {"weight": [[0.5625]]}}, "model.layers.69.mlp.down_proj": {"inputs": [[[7.8125]]], "outputs": [[[0.875]], [[1.1015625]]], "params": {"weight": [[0.4375]]}}, "model.layers.70.self_attn.qkv_proj": {"inputs": [[[12.625]]], "params": {"weight": [[0.435546875]]}}, "model.layers.70.self_attn.o_proj": {"inputs": [[[0.9609375]]], "outputs": [[[0.484375]], [[2.5]]], "params": {"weight": [[0.255859375]]}}, "model.layers.70.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.70703125]], [[14.1875]]]}, "model.layers.70.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.9921875]]]}, "model.layers.70.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[18.25]]]}, "model.layers.70.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.15625]]]}, "model.layers.70.self_attn.attn.impl.k_cache": {"inputs": [[[14.1875]]]}, "model.layers.70.self_attn.attn.impl.v_cache": {"inputs": [[[1.9921875]]]}, "model.layers.70.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.625]], [[12.875]], [[1.453125]]], "outputs": [[[0.85546875]], [[1.0]]]}, "model.layers.70.mlp.gate_up_proj": {"inputs": [[[8.875]]], "params": {"weight": [[0.53125]]}}, "model.layers.70.mlp.down_proj": {"inputs": [[[8.3125]]], "outputs": [[[1.625]], [[4.875]]], "params": {"weight": [[0.337890625]]}}, "model.layers.71.self_attn.qkv_proj": {"inputs": [[[13.0625]]], "params": {"weight": [[0.28125]]}}, "model.layers.71.self_attn.o_proj": {"inputs": [[[1.1875]]], "outputs": [[[1.125]], [[5.34375]]], "params": {"weight": [[0.11767578125]]}}, "model.layers.71.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.73828125]], [[16.125]]]}, "model.layers.71.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.890625]]]}, "model.layers.71.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[10.0625]]]}, "model.layers.71.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[7.4375]]]}, "model.layers.71.self_attn.attn.impl.k_cache": {"inputs": [[[16.125]]]}, "model.layers.71.self_attn.attn.impl.v_cache": {"inputs": [[[1.890625]]]}, "model.layers.71.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.90625]], [[13.5625]], [[1.890625]]], "outputs": [[[0.8671875]], [[1.0]]]}, "model.layers.71.mlp.gate_up_proj": {"inputs": [[[9.4375]]], "params": {"weight": [[0.46484375]]}}, "model.layers.71.mlp.down_proj": {"inputs": [[[9.9375]]], "outputs": [[[1.703125]], [[3.953125]]], "params": {"weight": [[0.328125]]}}, "model.layers.72.self_attn.qkv_proj": {"inputs": [[[11.6875]]], "params": {"weight": [[0.302734375]]}}, "model.layers.72.self_attn.o_proj": {"inputs": [[[1.8125]]], "outputs": [[[0.81640625]], [[3.484375]]], "params": {"weight": [[1.0]]}}, "model.layers.72.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.9375]], [[14.8125]]]}, "model.layers.72.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.484375]]]}, "model.layers.72.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[11.125]]]}, "model.layers.72.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.65625]]]}, "model.layers.72.self_attn.attn.impl.k_cache": {"inputs": [[[14.8125]]]}, "model.layers.72.self_attn.attn.impl.v_cache": {"inputs": [[[2.484375]]]}, "model.layers.72.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[10.3125]], [[12.8125]], [[2.46875]]], "outputs": [[[1.0625]], [[1.0]]]}, "model.layers.72.mlp.gate_up_proj": {"inputs": [[[9.75]]], "params": {"weight": [[0.5234375]]}}, "model.layers.72.mlp.down_proj": {"inputs": [[[9.5]]], "outputs": [[[1.40625]], [[2.53125]]], "params": {"weight": [[1.203125]]}}, "model.layers.73.self_attn.qkv_proj": {"inputs": [[[12.0]]], "params": {"weight": [[0.248046875]]}}, "model.layers.73.self_attn.o_proj": {"inputs": [[[1.4609375]]], "outputs": [[[2.40625]], [[3.78125]]], "params": {"weight": [[0.6953125]]}}, "model.layers.73.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.828125]], [[19.125]]]}, "model.layers.73.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.515625]]]}, "model.layers.73.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[25.625]]]}, "model.layers.73.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[11.125]]]}, "model.layers.73.self_attn.attn.impl.k_cache": {"inputs": [[[19.125]]]}, "model.layers.73.self_attn.attn.impl.v_cache": {"inputs": [[[2.515625]]]}, "model.layers.73.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.75]], [[18.375]], [[2.65625]]], "outputs": [[[1.1328125]], [[1.0]]]}, "model.layers.73.mlp.gate_up_proj": {"inputs": [[[9.9375]]], "params": {"weight": [[0.474609375]]}}, "model.layers.73.mlp.down_proj": {"inputs": [[[9.75]]], "outputs": [[[2.78125]], [[4.25]]], "params": {"weight": [[0.91796875]]}}, "model.layers.74.self_attn.qkv_proj": {"inputs": [[[11.625]]], "params": {"weight": [[0.1611328125]]}}, "model.layers.74.self_attn.o_proj": {"inputs": [[[3.40625]]], "outputs": [[[3.15625]], [[5.59375]]], "params": {"weight": [[0.29296875]]}}, "model.layers.74.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.76953125]], [[17.5]]]}, "model.layers.74.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[5.125]]]}, "model.layers.74.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[10.1875]]]}, "model.layers.74.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[6.625]]]}, "model.layers.74.self_attn.attn.impl.k_cache": {"inputs": [[[17.5]]]}, "model.layers.74.self_attn.attn.impl.v_cache": {"inputs": [[[5.125]]]}, "model.layers.74.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[10.875]], [[17.25]], [[3.265625]]], "outputs": [[[1.6484375]], [[1.0]]]}, "model.layers.74.mlp.gate_up_proj": {"inputs": [[[8.8125]]], "params": {"weight": [[0.27734375]]}}, "model.layers.74.mlp.down_proj": {"inputs": [[[11.5]]], "outputs": [[[2.53125]], [[6.9375]]], "params": {"weight": [[0.90234375]]}}, "model.layers.75.self_attn.qkv_proj": {"inputs": [[[11.5625]]], "params": {"weight": [[0.1904296875]]}}, "model.layers.75.self_attn.o_proj": {"inputs": [[[2.6875]]], "outputs": [[[1.25]], [[6.65625]]], "params": {"weight": [[0.431640625]]}}, "model.layers.75.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.0078125]], [[15.875]]]}, "model.layers.75.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[3.578125]]]}, "model.layers.75.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[18.75]]]}, "model.layers.75.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[11.125]]]}, "model.layers.75.self_attn.attn.impl.k_cache": {"inputs": [[[15.875]]]}, "model.layers.75.self_attn.attn.impl.v_cache": {"inputs": [[[3.578125]]]}, "model.layers.75.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[12.3125]], [[14.4375]], [[2.984375]]], "outputs": [[[2.15625]], [[1.0]]]}, "model.layers.75.mlp.gate_up_proj": {"inputs": [[[8.8125]]], "params": {"weight": [[0.427734375]]}}, "model.layers.75.mlp.down_proj": {"inputs": [[[10.5625]]], "outputs": [[[9.875]], [[12.4375]]], "params": {"weight": [[0.63671875]]}}, "model.layers.76.self_attn.qkv_proj": {"inputs": [[[9.0]]], "params": {"weight": [[0.263671875]]}}, "model.layers.76.self_attn.o_proj": {"inputs": [[[1.84375]]], "outputs": [[[3.03125]], [[5.3125]]], "params": {"weight": [[0.58984375]]}}, "model.layers.76.self_attn.attn.impl.matmul_qk": {"inputs": [[[1.296875]], [[14.4375]]]}, "model.layers.76.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.546875]]]}, "model.layers.76.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[22.5]]]}, "model.layers.76.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[10.1875]]]}, "model.layers.76.self_attn.attn.impl.k_cache": {"inputs": [[[14.4375]]]}, "model.layers.76.self_attn.attn.impl.v_cache": {"inputs": [[[2.546875]]]}, "model.layers.76.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[12.5]], [[13.125]], [[2.546875]]], "outputs": [[[1.6953125]], [[1.0]]]}, "model.layers.76.mlp.gate_up_proj": {"inputs": [[[9.375]]], "params": {"weight": [[0.38671875]]}}, "model.layers.76.mlp.down_proj": {"inputs": [[[8.75]]], "outputs": [[[2.078125]], [[4.90625]]], "params": {"weight": [[0.72265625]]}}, "model.layers.77.self_attn.qkv_proj": {"inputs": [[[9.375]]], "params": {"weight": [[0.287109375]]}}, "model.layers.77.self_attn.o_proj": {"inputs": [[[2.375]]], "outputs": [[[2.703125]], [[3.96875]]], "params": {"weight": [[0.31640625]]}}, "model.layers.77.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.9453125]], [[14.6875]]]}, "model.layers.77.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[4.59375]]]}, "model.layers.77.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[24.375]]]}, "model.layers.77.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[14.0]]]}, "model.layers.77.self_attn.attn.impl.k_cache": {"inputs": [[[14.6875]]]}, "model.layers.77.self_attn.attn.impl.v_cache": {"inputs": [[[4.59375]]]}, "model.layers.77.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[9.75]], [[13.3125]], [[4.3125]]], "outputs": [[[2.375]], [[1.0]]]}, "model.layers.77.mlp.gate_up_proj": {"inputs": [[[11.0625]]], "params": {"weight": [[0.345703125]]}}, "model.layers.77.mlp.down_proj": {"inputs": [[[13.6875]]], "outputs": [[[4.03125]], [[6.8125]]], "params": {"weight": [[0.50390625]]}}, "model.layers.78.self_attn.qkv_proj": {"inputs": [[[8.625]]], "params": {"weight": [[0.365234375]]}}, "model.layers.78.self_attn.o_proj": {"inputs": [[[1.7890625]]], "outputs": [[[2.75]], [[4.03125]]], "params": {"weight": [[0.283203125]]}}, "model.layers.78.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.921875]], [[13.5625]]]}, "model.layers.78.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[2.65625]]]}, "model.layers.78.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[49.0]]]}, "model.layers.78.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[16.125]]]}, "model.layers.78.self_attn.attn.impl.k_cache": {"inputs": [[[13.5625]]]}, "model.layers.78.self_attn.attn.impl.v_cache": {"inputs": [[[2.65625]]]}, "model.layers.78.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[12.1875]], [[11.3125]], [[2.5]]], "outputs": [[[1.5078125]], [[1.0]]]}, "model.layers.78.mlp.gate_up_proj": {"inputs": [[[14.0]]], "params": {"weight": [[0.6640625]]}}, "model.layers.78.mlp.down_proj": {"inputs": [[[82.5]]], "outputs": [[[86.5]], [[125.0]]], "params": {"weight": [[0.58203125]]}}, "model.layers.79.self_attn.qkv_proj": {"inputs": [[[9.375]]], "params": {"weight": [[0.4296875]]}}, "model.layers.79.self_attn.o_proj": {"inputs": [[[1.1484375]]], "outputs": [[[4.125]], [[5.15625]]], "params": {"weight": [[0.2412109375]]}}, "model.layers.79.self_attn.attn.impl.matmul_qk": {"inputs": [[[0.63671875]], [[13.125]]]}, "model.layers.79.self_attn.attn.impl.matmul_av": {"inputs": [[[1.0]], [[1.8671875]]]}, "model.layers.79.self_attn.attn.impl.batch2block_matmul": {"inputs": [[[1.0]], [[34.25]]]}, "model.layers.79.self_attn.attn.impl.block2batch_matmul": {"inputs": [[[1.0]], [[20.75]]]}, "model.layers.79.self_attn.attn.impl.k_cache": {"inputs": [[[13.125]]]}, "model.layers.79.self_attn.attn.impl.v_cache": {"inputs": [[[1.8671875]]]}, "model.layers.79.self_attn.attn.impl.fused_scaled_dot_product_attention": {"inputs": [[[7.125]], [[12.9375]], [[1.9609375]]], "outputs": [[[0.95703125]], [[1.0]]]}, "model.layers.79.mlp.gate_up_proj": {"inputs": [[[17.25]]], "params": {"weight": [[0.5078125]]}}, "model.layers.79.mlp.down_proj": {"inputs": [[[44.25]]], "outputs": [[[155.0]], [[402.0]]], "params": {"weight": [[0.6796875]]}}, "lm_head": {"inputs": [[[81.0]]], "params": {"weight": [[0.251953125]]}}}}

quant/g3/inc_output_hooks_maxabs_3_4.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8011c1c1bc44ac18ac30a7148905c6d88a49653d9609b3be41ce2425f9c917e
+size 206298

quant/g3/inc_output_hooks_maxabs_3_4_mod_list.json

@@ -0,0 +1,963 @@
+[
+    "model.layers.0.self_attn.qkv_proj",
+    "model.layers.0.self_attn.o_proj",
+    "model.layers.0.self_attn.attn.impl.matmul_qk",
+    "model.layers.0.self_attn.attn.impl.softmax",
+    "model.layers.0.self_attn.attn.impl.matmul_av",
+    "model.layers.0.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.0.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.0.self_attn.attn.impl.k_cache",
+    "model.layers.0.self_attn.attn.impl.v_cache",
+    "model.layers.0.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.0.mlp.gate_up_proj",
+    "model.layers.0.mlp.down_proj",
+    "model.layers.1.self_attn.qkv_proj",
+    "model.layers.1.self_attn.o_proj",
+    "model.layers.1.self_attn.attn.impl.matmul_qk",
+    "model.layers.1.self_attn.attn.impl.softmax",
+    "model.layers.1.self_attn.attn.impl.matmul_av",
+    "model.layers.1.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.1.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.1.self_attn.attn.impl.k_cache",
+    "model.layers.1.self_attn.attn.impl.v_cache",
+    "model.layers.1.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.1.mlp.gate_up_proj",
+    "model.layers.1.mlp.down_proj",
+    "model.layers.2.self_attn.qkv_proj",
+    "model.layers.2.self_attn.o_proj",
+    "model.layers.2.self_attn.attn.impl.matmul_qk",
+    "model.layers.2.self_attn.attn.impl.softmax",
+    "model.layers.2.self_attn.attn.impl.matmul_av",
+    "model.layers.2.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.2.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.2.self_attn.attn.impl.k_cache",
+    "model.layers.2.self_attn.attn.impl.v_cache",
+    "model.layers.2.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.2.mlp.gate_up_proj",
+    "model.layers.2.mlp.down_proj",
+    "model.layers.3.self_attn.qkv_proj",
+    "model.layers.3.self_attn.o_proj",
+    "model.layers.3.self_attn.attn.impl.matmul_qk",
+    "model.layers.3.self_attn.attn.impl.softmax",
+    "model.layers.3.self_attn.attn.impl.matmul_av",
+    "model.layers.3.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.3.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.3.self_attn.attn.impl.k_cache",
+    "model.layers.3.self_attn.attn.impl.v_cache",
+    "model.layers.3.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.3.mlp.gate_up_proj",
+    "model.layers.3.mlp.down_proj",
+    "model.layers.4.self_attn.qkv_proj",
+    "model.layers.4.self_attn.o_proj",
+    "model.layers.4.self_attn.attn.impl.matmul_qk",
+    "model.layers.4.self_attn.attn.impl.softmax",
+    "model.layers.4.self_attn.attn.impl.matmul_av",
+    "model.layers.4.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.4.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.4.self_attn.attn.impl.k_cache",
+    "model.layers.4.self_attn.attn.impl.v_cache",
+    "model.layers.4.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.4.mlp.gate_up_proj",
+    "model.layers.4.mlp.down_proj",
+    "model.layers.5.self_attn.qkv_proj",
+    "model.layers.5.self_attn.o_proj",
+    "model.layers.5.self_attn.attn.impl.matmul_qk",
+    "model.layers.5.self_attn.attn.impl.softmax",
+    "model.layers.5.self_attn.attn.impl.matmul_av",
+    "model.layers.5.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.5.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.5.self_attn.attn.impl.k_cache",
+    "model.layers.5.self_attn.attn.impl.v_cache",
+    "model.layers.5.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.5.mlp.gate_up_proj",
+    "model.layers.5.mlp.down_proj",
+    "model.layers.6.self_attn.qkv_proj",
+    "model.layers.6.self_attn.o_proj",
+    "model.layers.6.self_attn.attn.impl.matmul_qk",
+    "model.layers.6.self_attn.attn.impl.softmax",
+    "model.layers.6.self_attn.attn.impl.matmul_av",
+    "model.layers.6.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.6.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.6.self_attn.attn.impl.k_cache",
+    "model.layers.6.self_attn.attn.impl.v_cache",
+    "model.layers.6.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.6.mlp.gate_up_proj",
+    "model.layers.6.mlp.down_proj",
+    "model.layers.7.self_attn.qkv_proj",
+    "model.layers.7.self_attn.o_proj",
+    "model.layers.7.self_attn.attn.impl.matmul_qk",
+    "model.layers.7.self_attn.attn.impl.softmax",
+    "model.layers.7.self_attn.attn.impl.matmul_av",
+    "model.layers.7.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.7.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.7.self_attn.attn.impl.k_cache",
+    "model.layers.7.self_attn.attn.impl.v_cache",
+    "model.layers.7.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.7.mlp.gate_up_proj",
+    "model.layers.7.mlp.down_proj",
+    "model.layers.8.self_attn.qkv_proj",
+    "model.layers.8.self_attn.o_proj",
+    "model.layers.8.self_attn.attn.impl.matmul_qk",
+    "model.layers.8.self_attn.attn.impl.softmax",
+    "model.layers.8.self_attn.attn.impl.matmul_av",
+    "model.layers.8.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.8.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.8.self_attn.attn.impl.k_cache",
+    "model.layers.8.self_attn.attn.impl.v_cache",
+    "model.layers.8.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.8.mlp.gate_up_proj",
+    "model.layers.8.mlp.down_proj",
+    "model.layers.9.self_attn.qkv_proj",
+    "model.layers.9.self_attn.o_proj",
+    "model.layers.9.self_attn.attn.impl.matmul_qk",
+    "model.layers.9.self_attn.attn.impl.softmax",
+    "model.layers.9.self_attn.attn.impl.matmul_av",
+    "model.layers.9.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.9.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.9.self_attn.attn.impl.k_cache",
+    "model.layers.9.self_attn.attn.impl.v_cache",
+    "model.layers.9.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.9.mlp.gate_up_proj",
+    "model.layers.9.mlp.down_proj",
+    "model.layers.10.self_attn.qkv_proj",
+    "model.layers.10.self_attn.o_proj",
+    "model.layers.10.self_attn.attn.impl.matmul_qk",
+    "model.layers.10.self_attn.attn.impl.softmax",
+    "model.layers.10.self_attn.attn.impl.matmul_av",
+    "model.layers.10.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.10.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.10.self_attn.attn.impl.k_cache",
+    "model.layers.10.self_attn.attn.impl.v_cache",
+    "model.layers.10.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.10.mlp.gate_up_proj",
+    "model.layers.10.mlp.down_proj",
+    "model.layers.11.self_attn.qkv_proj",
+    "model.layers.11.self_attn.o_proj",
+    "model.layers.11.self_attn.attn.impl.matmul_qk",
+    "model.layers.11.self_attn.attn.impl.softmax",
+    "model.layers.11.self_attn.attn.impl.matmul_av",
+    "model.layers.11.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.11.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.11.self_attn.attn.impl.k_cache",
+    "model.layers.11.self_attn.attn.impl.v_cache",
+    "model.layers.11.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.11.mlp.gate_up_proj",
+    "model.layers.11.mlp.down_proj",
+    "model.layers.12.self_attn.qkv_proj",
+    "model.layers.12.self_attn.o_proj",
+    "model.layers.12.self_attn.attn.impl.matmul_qk",
+    "model.layers.12.self_attn.attn.impl.softmax",
+    "model.layers.12.self_attn.attn.impl.matmul_av",
+    "model.layers.12.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.12.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.12.self_attn.attn.impl.k_cache",
+    "model.layers.12.self_attn.attn.impl.v_cache",
+    "model.layers.12.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.12.mlp.gate_up_proj",
+    "model.layers.12.mlp.down_proj",
+    "model.layers.13.self_attn.qkv_proj",
+    "model.layers.13.self_attn.o_proj",
+    "model.layers.13.self_attn.attn.impl.matmul_qk",
+    "model.layers.13.self_attn.attn.impl.softmax",
+    "model.layers.13.self_attn.attn.impl.matmul_av",
+    "model.layers.13.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.13.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.13.self_attn.attn.impl.k_cache",
+    "model.layers.13.self_attn.attn.impl.v_cache",
+    "model.layers.13.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.13.mlp.gate_up_proj",
+    "model.layers.13.mlp.down_proj",
+    "model.layers.14.self_attn.qkv_proj",
+    "model.layers.14.self_attn.o_proj",
+    "model.layers.14.self_attn.attn.impl.matmul_qk",
+    "model.layers.14.self_attn.attn.impl.softmax",
+    "model.layers.14.self_attn.attn.impl.matmul_av",
+    "model.layers.14.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.14.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.14.self_attn.attn.impl.k_cache",
+    "model.layers.14.self_attn.attn.impl.v_cache",
+    "model.layers.14.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.14.mlp.gate_up_proj",
+    "model.layers.14.mlp.down_proj",
+    "model.layers.15.self_attn.qkv_proj",
+    "model.layers.15.self_attn.o_proj",
+    "model.layers.15.self_attn.attn.impl.matmul_qk",
+    "model.layers.15.self_attn.attn.impl.softmax",
+    "model.layers.15.self_attn.attn.impl.matmul_av",
+    "model.layers.15.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.15.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.15.self_attn.attn.impl.k_cache",
+    "model.layers.15.self_attn.attn.impl.v_cache",
+    "model.layers.15.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.15.mlp.gate_up_proj",
+    "model.layers.15.mlp.down_proj",
+    "model.layers.16.self_attn.qkv_proj",
+    "model.layers.16.self_attn.o_proj",
+    "model.layers.16.self_attn.attn.impl.matmul_qk",
+    "model.layers.16.self_attn.attn.impl.softmax",
+    "model.layers.16.self_attn.attn.impl.matmul_av",
+    "model.layers.16.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.16.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.16.self_attn.attn.impl.k_cache",
+    "model.layers.16.self_attn.attn.impl.v_cache",
+    "model.layers.16.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.16.mlp.gate_up_proj",
+    "model.layers.16.mlp.down_proj",
+    "model.layers.17.self_attn.qkv_proj",
+    "model.layers.17.self_attn.o_proj",
+    "model.layers.17.self_attn.attn.impl.matmul_qk",
+    "model.layers.17.self_attn.attn.impl.softmax",
+    "model.layers.17.self_attn.attn.impl.matmul_av",
+    "model.layers.17.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.17.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.17.self_attn.attn.impl.k_cache",
+    "model.layers.17.self_attn.attn.impl.v_cache",
+    "model.layers.17.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.17.mlp.gate_up_proj",
+    "model.layers.17.mlp.down_proj",
+    "model.layers.18.self_attn.qkv_proj",
+    "model.layers.18.self_attn.o_proj",
+    "model.layers.18.self_attn.attn.impl.matmul_qk",
+    "model.layers.18.self_attn.attn.impl.softmax",
+    "model.layers.18.self_attn.attn.impl.matmul_av",
+    "model.layers.18.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.18.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.18.self_attn.attn.impl.k_cache",
+    "model.layers.18.self_attn.attn.impl.v_cache",
+    "model.layers.18.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.18.mlp.gate_up_proj",
+    "model.layers.18.mlp.down_proj",
+    "model.layers.19.self_attn.qkv_proj",
+    "model.layers.19.self_attn.o_proj",
+    "model.layers.19.self_attn.attn.impl.matmul_qk",
+    "model.layers.19.self_attn.attn.impl.softmax",
+    "model.layers.19.self_attn.attn.impl.matmul_av",
+    "model.layers.19.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.19.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.19.self_attn.attn.impl.k_cache",
+    "model.layers.19.self_attn.attn.impl.v_cache",
+    "model.layers.19.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.19.mlp.gate_up_proj",
+    "model.layers.19.mlp.down_proj",
+    "model.layers.20.self_attn.qkv_proj",
+    "model.layers.20.self_attn.o_proj",
+    "model.layers.20.self_attn.attn.impl.matmul_qk",
+    "model.layers.20.self_attn.attn.impl.softmax",
+    "model.layers.20.self_attn.attn.impl.matmul_av",
+    "model.layers.20.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.20.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.20.self_attn.attn.impl.k_cache",
+    "model.layers.20.self_attn.attn.impl.v_cache",
+    "model.layers.20.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.20.mlp.gate_up_proj",
+    "model.layers.20.mlp.down_proj",
+    "model.layers.21.self_attn.qkv_proj",
+    "model.layers.21.self_attn.o_proj",
+    "model.layers.21.self_attn.attn.impl.matmul_qk",
+    "model.layers.21.self_attn.attn.impl.softmax",
+    "model.layers.21.self_attn.attn.impl.matmul_av",
+    "model.layers.21.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.21.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.21.self_attn.attn.impl.k_cache",
+    "model.layers.21.self_attn.attn.impl.v_cache",
+    "model.layers.21.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.21.mlp.gate_up_proj",
+    "model.layers.21.mlp.down_proj",
+    "model.layers.22.self_attn.qkv_proj",
+    "model.layers.22.self_attn.o_proj",
+    "model.layers.22.self_attn.attn.impl.matmul_qk",
+    "model.layers.22.self_attn.attn.impl.softmax",
+    "model.layers.22.self_attn.attn.impl.matmul_av",
+    "model.layers.22.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.22.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.22.self_attn.attn.impl.k_cache",
+    "model.layers.22.self_attn.attn.impl.v_cache",
+    "model.layers.22.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.22.mlp.gate_up_proj",
+    "model.layers.22.mlp.down_proj",
+    "model.layers.23.self_attn.qkv_proj",
+    "model.layers.23.self_attn.o_proj",
+    "model.layers.23.self_attn.attn.impl.matmul_qk",
+    "model.layers.23.self_attn.attn.impl.softmax",
+    "model.layers.23.self_attn.attn.impl.matmul_av",
+    "model.layers.23.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.23.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.23.self_attn.attn.impl.k_cache",
+    "model.layers.23.self_attn.attn.impl.v_cache",
+    "model.layers.23.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.23.mlp.gate_up_proj",
+    "model.layers.23.mlp.down_proj",
+    "model.layers.24.self_attn.qkv_proj",
+    "model.layers.24.self_attn.o_proj",
+    "model.layers.24.self_attn.attn.impl.matmul_qk",
+    "model.layers.24.self_attn.attn.impl.softmax",
+    "model.layers.24.self_attn.attn.impl.matmul_av",
+    "model.layers.24.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.24.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.24.self_attn.attn.impl.k_cache",
+    "model.layers.24.self_attn.attn.impl.v_cache",
+    "model.layers.24.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.24.mlp.gate_up_proj",
+    "model.layers.24.mlp.down_proj",
+    "model.layers.25.self_attn.qkv_proj",
+    "model.layers.25.self_attn.o_proj",
+    "model.layers.25.self_attn.attn.impl.matmul_qk",
+    "model.layers.25.self_attn.attn.impl.softmax",
+    "model.layers.25.self_attn.attn.impl.matmul_av",
+    "model.layers.25.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.25.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.25.self_attn.attn.impl.k_cache",
+    "model.layers.25.self_attn.attn.impl.v_cache",
+    "model.layers.25.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.25.mlp.gate_up_proj",
+    "model.layers.25.mlp.down_proj",
+    "model.layers.26.self_attn.qkv_proj",
+    "model.layers.26.self_attn.o_proj",
+    "model.layers.26.self_attn.attn.impl.matmul_qk",
+    "model.layers.26.self_attn.attn.impl.softmax",
+    "model.layers.26.self_attn.attn.impl.matmul_av",
+    "model.layers.26.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.26.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.26.self_attn.attn.impl.k_cache",
+    "model.layers.26.self_attn.attn.impl.v_cache",
+    "model.layers.26.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.26.mlp.gate_up_proj",
+    "model.layers.26.mlp.down_proj",
+    "model.layers.27.self_attn.qkv_proj",
+    "model.layers.27.self_attn.o_proj",
+    "model.layers.27.self_attn.attn.impl.matmul_qk",
+    "model.layers.27.self_attn.attn.impl.softmax",
+    "model.layers.27.self_attn.attn.impl.matmul_av",
+    "model.layers.27.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.27.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.27.self_attn.attn.impl.k_cache",
+    "model.layers.27.self_attn.attn.impl.v_cache",
+    "model.layers.27.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.27.mlp.gate_up_proj",
+    "model.layers.27.mlp.down_proj",
+    "model.layers.28.self_attn.qkv_proj",
+    "model.layers.28.self_attn.o_proj",
+    "model.layers.28.self_attn.attn.impl.matmul_qk",
+    "model.layers.28.self_attn.attn.impl.softmax",
+    "model.layers.28.self_attn.attn.impl.matmul_av",
+    "model.layers.28.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.28.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.28.self_attn.attn.impl.k_cache",
+    "model.layers.28.self_attn.attn.impl.v_cache",
+    "model.layers.28.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.28.mlp.gate_up_proj",
+    "model.layers.28.mlp.down_proj",
+    "model.layers.29.self_attn.qkv_proj",
+    "model.layers.29.self_attn.o_proj",
+    "model.layers.29.self_attn.attn.impl.matmul_qk",
+    "model.layers.29.self_attn.attn.impl.softmax",
+    "model.layers.29.self_attn.attn.impl.matmul_av",
+    "model.layers.29.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.29.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.29.self_attn.attn.impl.k_cache",
+    "model.layers.29.self_attn.attn.impl.v_cache",
+    "model.layers.29.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.29.mlp.gate_up_proj",
+    "model.layers.29.mlp.down_proj",
+    "model.layers.30.self_attn.qkv_proj",
+    "model.layers.30.self_attn.o_proj",
+    "model.layers.30.self_attn.attn.impl.matmul_qk",
+    "model.layers.30.self_attn.attn.impl.softmax",
+    "model.layers.30.self_attn.attn.impl.matmul_av",
+    "model.layers.30.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.30.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.30.self_attn.attn.impl.k_cache",
+    "model.layers.30.self_attn.attn.impl.v_cache",
+    "model.layers.30.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.30.mlp.gate_up_proj",
+    "model.layers.30.mlp.down_proj",
+    "model.layers.31.self_attn.qkv_proj",
+    "model.layers.31.self_attn.o_proj",
+    "model.layers.31.self_attn.attn.impl.matmul_qk",
+    "model.layers.31.self_attn.attn.impl.softmax",
+    "model.layers.31.self_attn.attn.impl.matmul_av",
+    "model.layers.31.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.31.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.31.self_attn.attn.impl.k_cache",
+    "model.layers.31.self_attn.attn.impl.v_cache",
+    "model.layers.31.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.31.mlp.gate_up_proj",
+    "model.layers.31.mlp.down_proj",
+    "model.layers.32.self_attn.qkv_proj",
+    "model.layers.32.self_attn.o_proj",
+    "model.layers.32.self_attn.attn.impl.matmul_qk",
+    "model.layers.32.self_attn.attn.impl.softmax",
+    "model.layers.32.self_attn.attn.impl.matmul_av",
+    "model.layers.32.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.32.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.32.self_attn.attn.impl.k_cache",
+    "model.layers.32.self_attn.attn.impl.v_cache",
+    "model.layers.32.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.32.mlp.gate_up_proj",
+    "model.layers.32.mlp.down_proj",
+    "model.layers.33.self_attn.qkv_proj",
+    "model.layers.33.self_attn.o_proj",
+    "model.layers.33.self_attn.attn.impl.matmul_qk",
+    "model.layers.33.self_attn.attn.impl.softmax",
+    "model.layers.33.self_attn.attn.impl.matmul_av",
+    "model.layers.33.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.33.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.33.self_attn.attn.impl.k_cache",
+    "model.layers.33.self_attn.attn.impl.v_cache",
+    "model.layers.33.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.33.mlp.gate_up_proj",
+    "model.layers.33.mlp.down_proj",
+    "model.layers.34.self_attn.qkv_proj",
+    "model.layers.34.self_attn.o_proj",
+    "model.layers.34.self_attn.attn.impl.matmul_qk",
+    "model.layers.34.self_attn.attn.impl.softmax",
+    "model.layers.34.self_attn.attn.impl.matmul_av",
+    "model.layers.34.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.34.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.34.self_attn.attn.impl.k_cache",
+    "model.layers.34.self_attn.attn.impl.v_cache",
+    "model.layers.34.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.34.mlp.gate_up_proj",
+    "model.layers.34.mlp.down_proj",
+    "model.layers.35.self_attn.qkv_proj",
+    "model.layers.35.self_attn.o_proj",
+    "model.layers.35.self_attn.attn.impl.matmul_qk",
+    "model.layers.35.self_attn.attn.impl.softmax",
+    "model.layers.35.self_attn.attn.impl.matmul_av",
+    "model.layers.35.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.35.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.35.self_attn.attn.impl.k_cache",
+    "model.layers.35.self_attn.attn.impl.v_cache",
+    "model.layers.35.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.35.mlp.gate_up_proj",
+    "model.layers.35.mlp.down_proj",
+    "model.layers.36.self_attn.qkv_proj",
+    "model.layers.36.self_attn.o_proj",
+    "model.layers.36.self_attn.attn.impl.matmul_qk",
+    "model.layers.36.self_attn.attn.impl.softmax",
+    "model.layers.36.self_attn.attn.impl.matmul_av",
+    "model.layers.36.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.36.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.36.self_attn.attn.impl.k_cache",
+    "model.layers.36.self_attn.attn.impl.v_cache",
+    "model.layers.36.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.36.mlp.gate_up_proj",
+    "model.layers.36.mlp.down_proj",
+    "model.layers.37.self_attn.qkv_proj",
+    "model.layers.37.self_attn.o_proj",
+    "model.layers.37.self_attn.attn.impl.matmul_qk",
+    "model.layers.37.self_attn.attn.impl.softmax",
+    "model.layers.37.self_attn.attn.impl.matmul_av",
+    "model.layers.37.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.37.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.37.self_attn.attn.impl.k_cache",
+    "model.layers.37.self_attn.attn.impl.v_cache",
+    "model.layers.37.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.37.mlp.gate_up_proj",
+    "model.layers.37.mlp.down_proj",
+    "model.layers.38.self_attn.qkv_proj",
+    "model.layers.38.self_attn.o_proj",
+    "model.layers.38.self_attn.attn.impl.matmul_qk",
+    "model.layers.38.self_attn.attn.impl.softmax",
+    "model.layers.38.self_attn.attn.impl.matmul_av",
+    "model.layers.38.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.38.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.38.self_attn.attn.impl.k_cache",
+    "model.layers.38.self_attn.attn.impl.v_cache",
+    "model.layers.38.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.38.mlp.gate_up_proj",
+    "model.layers.38.mlp.down_proj",
+    "model.layers.39.self_attn.qkv_proj",
+    "model.layers.39.self_attn.o_proj",
+    "model.layers.39.self_attn.attn.impl.matmul_qk",
+    "model.layers.39.self_attn.attn.impl.softmax",
+    "model.layers.39.self_attn.attn.impl.matmul_av",
+    "model.layers.39.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.39.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.39.self_attn.attn.impl.k_cache",
+    "model.layers.39.self_attn.attn.impl.v_cache",
+    "model.layers.39.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.39.mlp.gate_up_proj",
+    "model.layers.39.mlp.down_proj",
+    "model.layers.40.self_attn.qkv_proj",
+    "model.layers.40.self_attn.o_proj",
+    "model.layers.40.self_attn.attn.impl.matmul_qk",
+    "model.layers.40.self_attn.attn.impl.softmax",
+    "model.layers.40.self_attn.attn.impl.matmul_av",
+    "model.layers.40.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.40.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.40.self_attn.attn.impl.k_cache",
+    "model.layers.40.self_attn.attn.impl.v_cache",
+    "model.layers.40.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.40.mlp.gate_up_proj",
+    "model.layers.40.mlp.down_proj",
+    "model.layers.41.self_attn.qkv_proj",
+    "model.layers.41.self_attn.o_proj",
+    "model.layers.41.self_attn.attn.impl.matmul_qk",
+    "model.layers.41.self_attn.attn.impl.softmax",
+    "model.layers.41.self_attn.attn.impl.matmul_av",
+    "model.layers.41.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.41.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.41.self_attn.attn.impl.k_cache",
+    "model.layers.41.self_attn.attn.impl.v_cache",
+    "model.layers.41.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.41.mlp.gate_up_proj",
+    "model.layers.41.mlp.down_proj",
+    "model.layers.42.self_attn.qkv_proj",
+    "model.layers.42.self_attn.o_proj",
+    "model.layers.42.self_attn.attn.impl.matmul_qk",
+    "model.layers.42.self_attn.attn.impl.softmax",
+    "model.layers.42.self_attn.attn.impl.matmul_av",
+    "model.layers.42.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.42.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.42.self_attn.attn.impl.k_cache",
+    "model.layers.42.self_attn.attn.impl.v_cache",
+    "model.layers.42.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.42.mlp.gate_up_proj",
+    "model.layers.42.mlp.down_proj",
+    "model.layers.43.self_attn.qkv_proj",
+    "model.layers.43.self_attn.o_proj",
+    "model.layers.43.self_attn.attn.impl.matmul_qk",
+    "model.layers.43.self_attn.attn.impl.softmax",
+    "model.layers.43.self_attn.attn.impl.matmul_av",
+    "model.layers.43.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.43.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.43.self_attn.attn.impl.k_cache",
+    "model.layers.43.self_attn.attn.impl.v_cache",
+    "model.layers.43.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.43.mlp.gate_up_proj",
+    "model.layers.43.mlp.down_proj",
+    "model.layers.44.self_attn.qkv_proj",
+    "model.layers.44.self_attn.o_proj",
+    "model.layers.44.self_attn.attn.impl.matmul_qk",
+    "model.layers.44.self_attn.attn.impl.softmax",
+    "model.layers.44.self_attn.attn.impl.matmul_av",
+    "model.layers.44.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.44.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.44.self_attn.attn.impl.k_cache",
+    "model.layers.44.self_attn.attn.impl.v_cache",
+    "model.layers.44.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.44.mlp.gate_up_proj",
+    "model.layers.44.mlp.down_proj",
+    "model.layers.45.self_attn.qkv_proj",
+    "model.layers.45.self_attn.o_proj",
+    "model.layers.45.self_attn.attn.impl.matmul_qk",
+    "model.layers.45.self_attn.attn.impl.softmax",
+    "model.layers.45.self_attn.attn.impl.matmul_av",
+    "model.layers.45.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.45.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.45.self_attn.attn.impl.k_cache",
+    "model.layers.45.self_attn.attn.impl.v_cache",
+    "model.layers.45.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.45.mlp.gate_up_proj",
+    "model.layers.45.mlp.down_proj",
+    "model.layers.46.self_attn.qkv_proj",
+    "model.layers.46.self_attn.o_proj",
+    "model.layers.46.self_attn.attn.impl.matmul_qk",
+    "model.layers.46.self_attn.attn.impl.softmax",
+    "model.layers.46.self_attn.attn.impl.matmul_av",
+    "model.layers.46.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.46.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.46.self_attn.attn.impl.k_cache",
+    "model.layers.46.self_attn.attn.impl.v_cache",
+    "model.layers.46.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.46.mlp.gate_up_proj",
+    "model.layers.46.mlp.down_proj",
+    "model.layers.47.self_attn.qkv_proj",
+    "model.layers.47.self_attn.o_proj",
+    "model.layers.47.self_attn.attn.impl.matmul_qk",
+    "model.layers.47.self_attn.attn.impl.softmax",
+    "model.layers.47.self_attn.attn.impl.matmul_av",
+    "model.layers.47.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.47.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.47.self_attn.attn.impl.k_cache",
+    "model.layers.47.self_attn.attn.impl.v_cache",
+    "model.layers.47.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.47.mlp.gate_up_proj",
+    "model.layers.47.mlp.down_proj",
+    "model.layers.48.self_attn.qkv_proj",
+    "model.layers.48.self_attn.o_proj",
+    "model.layers.48.self_attn.attn.impl.matmul_qk",
+    "model.layers.48.self_attn.attn.impl.softmax",
+    "model.layers.48.self_attn.attn.impl.matmul_av",
+    "model.layers.48.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.48.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.48.self_attn.attn.impl.k_cache",
+    "model.layers.48.self_attn.attn.impl.v_cache",
+    "model.layers.48.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.48.mlp.gate_up_proj",
+    "model.layers.48.mlp.down_proj",
+    "model.layers.49.self_attn.qkv_proj",
+    "model.layers.49.self_attn.o_proj",
+    "model.layers.49.self_attn.attn.impl.matmul_qk",
+    "model.layers.49.self_attn.attn.impl.softmax",
+    "model.layers.49.self_attn.attn.impl.matmul_av",
+    "model.layers.49.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.49.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.49.self_attn.attn.impl.k_cache",
+    "model.layers.49.self_attn.attn.impl.v_cache",
+    "model.layers.49.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.49.mlp.gate_up_proj",
+    "model.layers.49.mlp.down_proj",
+    "model.layers.50.self_attn.qkv_proj",
+    "model.layers.50.self_attn.o_proj",
+    "model.layers.50.self_attn.attn.impl.matmul_qk",
+    "model.layers.50.self_attn.attn.impl.softmax",
+    "model.layers.50.self_attn.attn.impl.matmul_av",
+    "model.layers.50.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.50.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.50.self_attn.attn.impl.k_cache",
+    "model.layers.50.self_attn.attn.impl.v_cache",
+    "model.layers.50.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.50.mlp.gate_up_proj",
+    "model.layers.50.mlp.down_proj",
+    "model.layers.51.self_attn.qkv_proj",
+    "model.layers.51.self_attn.o_proj",
+    "model.layers.51.self_attn.attn.impl.matmul_qk",
+    "model.layers.51.self_attn.attn.impl.softmax",
+    "model.layers.51.self_attn.attn.impl.matmul_av",
+    "model.layers.51.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.51.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.51.self_attn.attn.impl.k_cache",
+    "model.layers.51.self_attn.attn.impl.v_cache",
+    "model.layers.51.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.51.mlp.gate_up_proj",
+    "model.layers.51.mlp.down_proj",
+    "model.layers.52.self_attn.qkv_proj",
+    "model.layers.52.self_attn.o_proj",
+    "model.layers.52.self_attn.attn.impl.matmul_qk",
+    "model.layers.52.self_attn.attn.impl.softmax",
+    "model.layers.52.self_attn.attn.impl.matmul_av",
+    "model.layers.52.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.52.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.52.self_attn.attn.impl.k_cache",
+    "model.layers.52.self_attn.attn.impl.v_cache",
+    "model.layers.52.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.52.mlp.gate_up_proj",
+    "model.layers.52.mlp.down_proj",
+    "model.layers.53.self_attn.qkv_proj",
+    "model.layers.53.self_attn.o_proj",
+    "model.layers.53.self_attn.attn.impl.matmul_qk",
+    "model.layers.53.self_attn.attn.impl.softmax",
+    "model.layers.53.self_attn.attn.impl.matmul_av",
+    "model.layers.53.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.53.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.53.self_attn.attn.impl.k_cache",
+    "model.layers.53.self_attn.attn.impl.v_cache",
+    "model.layers.53.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.53.mlp.gate_up_proj",
+    "model.layers.53.mlp.down_proj",
+    "model.layers.54.self_attn.qkv_proj",
+    "model.layers.54.self_attn.o_proj",
+    "model.layers.54.self_attn.attn.impl.matmul_qk",
+    "model.layers.54.self_attn.attn.impl.softmax",
+    "model.layers.54.self_attn.attn.impl.matmul_av",
+    "model.layers.54.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.54.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.54.self_attn.attn.impl.k_cache",
+    "model.layers.54.self_attn.attn.impl.v_cache",
+    "model.layers.54.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.54.mlp.gate_up_proj",
+    "model.layers.54.mlp.down_proj",
+    "model.layers.55.self_attn.qkv_proj",
+    "model.layers.55.self_attn.o_proj",
+    "model.layers.55.self_attn.attn.impl.matmul_qk",
+    "model.layers.55.self_attn.attn.impl.softmax",
+    "model.layers.55.self_attn.attn.impl.matmul_av",
+    "model.layers.55.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.55.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.55.self_attn.attn.impl.k_cache",
+    "model.layers.55.self_attn.attn.impl.v_cache",
+    "model.layers.55.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.55.mlp.gate_up_proj",
+    "model.layers.55.mlp.down_proj",
+    "model.layers.56.self_attn.qkv_proj",
+    "model.layers.56.self_attn.o_proj",
+    "model.layers.56.self_attn.attn.impl.matmul_qk",
+    "model.layers.56.self_attn.attn.impl.softmax",
+    "model.layers.56.self_attn.attn.impl.matmul_av",
+    "model.layers.56.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.56.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.56.self_attn.attn.impl.k_cache",
+    "model.layers.56.self_attn.attn.impl.v_cache",
+    "model.layers.56.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.56.mlp.gate_up_proj",
+    "model.layers.56.mlp.down_proj",
+    "model.layers.57.self_attn.qkv_proj",
+    "model.layers.57.self_attn.o_proj",
+    "model.layers.57.self_attn.attn.impl.matmul_qk",
+    "model.layers.57.self_attn.attn.impl.softmax",
+    "model.layers.57.self_attn.attn.impl.matmul_av",
+    "model.layers.57.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.57.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.57.self_attn.attn.impl.k_cache",
+    "model.layers.57.self_attn.attn.impl.v_cache",
+    "model.layers.57.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.57.mlp.gate_up_proj",
+    "model.layers.57.mlp.down_proj",
+    "model.layers.58.self_attn.qkv_proj",
+    "model.layers.58.self_attn.o_proj",
+    "model.layers.58.self_attn.attn.impl.matmul_qk",
+    "model.layers.58.self_attn.attn.impl.softmax",
+    "model.layers.58.self_attn.attn.impl.matmul_av",
+    "model.layers.58.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.58.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.58.self_attn.attn.impl.k_cache",
+    "model.layers.58.self_attn.attn.impl.v_cache",
+    "model.layers.58.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.58.mlp.gate_up_proj",
+    "model.layers.58.mlp.down_proj",
+    "model.layers.59.self_attn.qkv_proj",
+    "model.layers.59.self_attn.o_proj",
+    "model.layers.59.self_attn.attn.impl.matmul_qk",
+    "model.layers.59.self_attn.attn.impl.softmax",
+    "model.layers.59.self_attn.attn.impl.matmul_av",
+    "model.layers.59.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.59.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.59.self_attn.attn.impl.k_cache",
+    "model.layers.59.self_attn.attn.impl.v_cache",
+    "model.layers.59.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.59.mlp.gate_up_proj",
+    "model.layers.59.mlp.down_proj",
+    "model.layers.60.self_attn.qkv_proj",
+    "model.layers.60.self_attn.o_proj",
+    "model.layers.60.self_attn.attn.impl.matmul_qk",
+    "model.layers.60.self_attn.attn.impl.softmax",
+    "model.layers.60.self_attn.attn.impl.matmul_av",
+    "model.layers.60.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.60.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.60.self_attn.attn.impl.k_cache",
+    "model.layers.60.self_attn.attn.impl.v_cache",
+    "model.layers.60.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.60.mlp.gate_up_proj",
+    "model.layers.60.mlp.down_proj",
+    "model.layers.61.self_attn.qkv_proj",
+    "model.layers.61.self_attn.o_proj",
+    "model.layers.61.self_attn.attn.impl.matmul_qk",
+    "model.layers.61.self_attn.attn.impl.softmax",
+    "model.layers.61.self_attn.attn.impl.matmul_av",
+    "model.layers.61.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.61.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.61.self_attn.attn.impl.k_cache",
+    "model.layers.61.self_attn.attn.impl.v_cache",
+    "model.layers.61.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.61.mlp.gate_up_proj",
+    "model.layers.61.mlp.down_proj",
+    "model.layers.62.self_attn.qkv_proj",
+    "model.layers.62.self_attn.o_proj",
+    "model.layers.62.self_attn.attn.impl.matmul_qk",
+    "model.layers.62.self_attn.attn.impl.softmax",
+    "model.layers.62.self_attn.attn.impl.matmul_av",
+    "model.layers.62.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.62.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.62.self_attn.attn.impl.k_cache",
+    "model.layers.62.self_attn.attn.impl.v_cache",
+    "model.layers.62.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.62.mlp.gate_up_proj",
+    "model.layers.62.mlp.down_proj",
+    "model.layers.63.self_attn.qkv_proj",
+    "model.layers.63.self_attn.o_proj",
+    "model.layers.63.self_attn.attn.impl.matmul_qk",
+    "model.layers.63.self_attn.attn.impl.softmax",
+    "model.layers.63.self_attn.attn.impl.matmul_av",
+    "model.layers.63.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.63.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.63.self_attn.attn.impl.k_cache",
+    "model.layers.63.self_attn.attn.impl.v_cache",
+    "model.layers.63.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.63.mlp.gate_up_proj",
+    "model.layers.63.mlp.down_proj",
+    "model.layers.64.self_attn.qkv_proj",
+    "model.layers.64.self_attn.o_proj",
+    "model.layers.64.self_attn.attn.impl.matmul_qk",
+    "model.layers.64.self_attn.attn.impl.softmax",
+    "model.layers.64.self_attn.attn.impl.matmul_av",
+    "model.layers.64.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.64.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.64.self_attn.attn.impl.k_cache",
+    "model.layers.64.self_attn.attn.impl.v_cache",
+    "model.layers.64.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.64.mlp.gate_up_proj",
+    "model.layers.64.mlp.down_proj",
+    "model.layers.65.self_attn.qkv_proj",
+    "model.layers.65.self_attn.o_proj",
+    "model.layers.65.self_attn.attn.impl.matmul_qk",
+    "model.layers.65.self_attn.attn.impl.softmax",
+    "model.layers.65.self_attn.attn.impl.matmul_av",
+    "model.layers.65.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.65.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.65.self_attn.attn.impl.k_cache",
+    "model.layers.65.self_attn.attn.impl.v_cache",
+    "model.layers.65.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.65.mlp.gate_up_proj",
+    "model.layers.65.mlp.down_proj",
+    "model.layers.66.self_attn.qkv_proj",
+    "model.layers.66.self_attn.o_proj",
+    "model.layers.66.self_attn.attn.impl.matmul_qk",
+    "model.layers.66.self_attn.attn.impl.softmax",
+    "model.layers.66.self_attn.attn.impl.matmul_av",
+    "model.layers.66.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.66.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.66.self_attn.attn.impl.k_cache",
+    "model.layers.66.self_attn.attn.impl.v_cache",
+    "model.layers.66.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.66.mlp.gate_up_proj",
+    "model.layers.66.mlp.down_proj",
+    "model.layers.67.self_attn.qkv_proj",
+    "model.layers.67.self_attn.o_proj",
+    "model.layers.67.self_attn.attn.impl.matmul_qk",
+    "model.layers.67.self_attn.attn.impl.softmax",
+    "model.layers.67.self_attn.attn.impl.matmul_av",
+    "model.layers.67.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.67.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.67.self_attn.attn.impl.k_cache",
+    "model.layers.67.self_attn.attn.impl.v_cache",
+    "model.layers.67.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.67.mlp.gate_up_proj",
+    "model.layers.67.mlp.down_proj",
+    "model.layers.68.self_attn.qkv_proj",
+    "model.layers.68.self_attn.o_proj",
+    "model.layers.68.self_attn.attn.impl.matmul_qk",
+    "model.layers.68.self_attn.attn.impl.softmax",
+    "model.layers.68.self_attn.attn.impl.matmul_av",
+    "model.layers.68.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.68.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.68.self_attn.attn.impl.k_cache",
+    "model.layers.68.self_attn.attn.impl.v_cache",
+    "model.layers.68.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.68.mlp.gate_up_proj",
+    "model.layers.68.mlp.down_proj",
+    "model.layers.69.self_attn.qkv_proj",
+    "model.layers.69.self_attn.o_proj",
+    "model.layers.69.self_attn.attn.impl.matmul_qk",
+    "model.layers.69.self_attn.attn.impl.softmax",
+    "model.layers.69.self_attn.attn.impl.matmul_av",
+    "model.layers.69.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.69.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.69.self_attn.attn.impl.k_cache",
+    "model.layers.69.self_attn.attn.impl.v_cache",
+    "model.layers.69.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.69.mlp.gate_up_proj",
+    "model.layers.69.mlp.down_proj",
+    "model.layers.70.self_attn.qkv_proj",
+    "model.layers.70.self_attn.o_proj",
+    "model.layers.70.self_attn.attn.impl.matmul_qk",
+    "model.layers.70.self_attn.attn.impl.softmax",
+    "model.layers.70.self_attn.attn.impl.matmul_av",
+    "model.layers.70.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.70.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.70.self_attn.attn.impl.k_cache",
+    "model.layers.70.self_attn.attn.impl.v_cache",
+    "model.layers.70.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.70.mlp.gate_up_proj",
+    "model.layers.70.mlp.down_proj",
+    "model.layers.71.self_attn.qkv_proj",
+    "model.layers.71.self_attn.o_proj",
+    "model.layers.71.self_attn.attn.impl.matmul_qk",
+    "model.layers.71.self_attn.attn.impl.softmax",
+    "model.layers.71.self_attn.attn.impl.matmul_av",
+    "model.layers.71.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.71.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.71.self_attn.attn.impl.k_cache",
+    "model.layers.71.self_attn.attn.impl.v_cache",
+    "model.layers.71.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.71.mlp.gate_up_proj",
+    "model.layers.71.mlp.down_proj",
+    "model.layers.72.self_attn.qkv_proj",
+    "model.layers.72.self_attn.o_proj",
+    "model.layers.72.self_attn.attn.impl.matmul_qk",
+    "model.layers.72.self_attn.attn.impl.softmax",
+    "model.layers.72.self_attn.attn.impl.matmul_av",
+    "model.layers.72.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.72.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.72.self_attn.attn.impl.k_cache",
+    "model.layers.72.self_attn.attn.impl.v_cache",
+    "model.layers.72.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.72.mlp.gate_up_proj",
+    "model.layers.72.mlp.down_proj",
+    "model.layers.73.self_attn.qkv_proj",
+    "model.layers.73.self_attn.o_proj",
+    "model.layers.73.self_attn.attn.impl.matmul_qk",
+    "model.layers.73.self_attn.attn.impl.softmax",
+    "model.layers.73.self_attn.attn.impl.matmul_av",
+    "model.layers.73.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.73.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.73.self_attn.attn.impl.k_cache",
+    "model.layers.73.self_attn.attn.impl.v_cache",
+    "model.layers.73.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.73.mlp.gate_up_proj",
+    "model.layers.73.mlp.down_proj",
+    "model.layers.74.self_attn.qkv_proj",
+    "model.layers.74.self_attn.o_proj",
+    "model.layers.74.self_attn.attn.impl.matmul_qk",
+    "model.layers.74.self_attn.attn.impl.softmax",
+    "model.layers.74.self_attn.attn.impl.matmul_av",
+    "model.layers.74.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.74.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.74.self_attn.attn.impl.k_cache",
+    "model.layers.74.self_attn.attn.impl.v_cache",
+    "model.layers.74.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.74.mlp.gate_up_proj",
+    "model.layers.74.mlp.down_proj",
+    "model.layers.75.self_attn.qkv_proj",
+    "model.layers.75.self_attn.o_proj",
+    "model.layers.75.self_attn.attn.impl.matmul_qk",
+    "model.layers.75.self_attn.attn.impl.softmax",
+    "model.layers.75.self_attn.attn.impl.matmul_av",
+    "model.layers.75.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.75.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.75.self_attn.attn.impl.k_cache",
+    "model.layers.75.self_attn.attn.impl.v_cache",
+    "model.layers.75.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.75.mlp.gate_up_proj",
+    "model.layers.75.mlp.down_proj",
+    "model.layers.76.self_attn.qkv_proj",
+    "model.layers.76.self_attn.o_proj",
+    "model.layers.76.self_attn.attn.impl.matmul_qk",
+    "model.layers.76.self_attn.attn.impl.softmax",
+    "model.layers.76.self_attn.attn.impl.matmul_av",
+    "model.layers.76.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.76.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.76.self_attn.attn.impl.k_cache",
+    "model.layers.76.self_attn.attn.impl.v_cache",
+    "model.layers.76.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.76.mlp.gate_up_proj",
+    "model.layers.76.mlp.down_proj",
+    "model.layers.77.self_attn.qkv_proj",
+    "model.layers.77.self_attn.o_proj",
+    "model.layers.77.self_attn.attn.impl.matmul_qk",
+    "model.layers.77.self_attn.attn.impl.softmax",
+    "model.layers.77.self_attn.attn.impl.matmul_av",
+    "model.layers.77.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.77.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.77.self_attn.attn.impl.k_cache",
+    "model.layers.77.self_attn.attn.impl.v_cache",
+    "model.layers.77.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.77.mlp.gate_up_proj",
+    "model.layers.77.mlp.down_proj",
+    "model.layers.78.self_attn.qkv_proj",
+    "model.layers.78.self_attn.o_proj",
+    "model.layers.78.self_attn.attn.impl.matmul_qk",
+    "model.layers.78.self_attn.attn.impl.softmax",
+    "model.layers.78.self_attn.attn.impl.matmul_av",
+    "model.layers.78.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.78.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.78.self_attn.attn.impl.k_cache",
+    "model.layers.78.self_attn.attn.impl.v_cache",
+    "model.layers.78.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.78.mlp.gate_up_proj",
+    "model.layers.78.mlp.down_proj",
+    "model.layers.79.self_attn.qkv_proj",
+    "model.layers.79.self_attn.o_proj",
+    "model.layers.79.self_attn.attn.impl.matmul_qk",
+    "model.layers.79.self_attn.attn.impl.softmax",
+    "model.layers.79.self_attn.attn.impl.matmul_av",
+    "model.layers.79.self_attn.attn.impl.batch2block_matmul",
+    "model.layers.79.self_attn.attn.impl.block2batch_matmul",
+    "model.layers.79.self_attn.attn.impl.k_cache",
+    "model.layers.79.self_attn.attn.impl.v_cache",
+    "model.layers.79.self_attn.attn.impl.fused_scaled_dot_product_attention",
+    "model.layers.79.mlp.gate_up_proj",
+    "model.layers.79.mlp.down_proj",
+    "lm_head"
+]

quant/g3/inc_output_hooks_maxabs_MAXABS_HW_0_4.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28fa12b482230e0d765930aafc5fba653d6b91b841c13d05f9164ead90cf96c7
+size 212484

quant/g3/inc_output_hooks_maxabs_MAXABS_HW_1_4.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8167b2ca512756adc9b4e6968ec6f2cdc4c3afc0716a857d454d11164991cff5
+size 212484

quant/g3/inc_output_hooks_maxabs_MAXABS_HW_2_4.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:898204590d3b3131975f00346d70a2a9dbaee039786bb635a8c76cda9592e38b
+size 212484

quant/g3/inc_output_hooks_maxabs_MAXABS_HW_3_4.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d9fc1daf3e06aa809f4c006efdbb55f1e36d66f09ab9a635223ad2a27d7a904
+size 212484

quant/maxabs_measure_g3.json

@@ -0,0 +1 @@
+{"method": "HOOKS","mode": "MEASURE","observer": "maxabs","allowlist": {"types": [], "names":  []},"blocklist": {"types": [], "names":  []},"quantize_weight": false,"dump_stats_path": "/software/ae/fmwork/inc/1.21.0/llama-3.3-70b-instruct/g3/inc_output"}

quant/maxabs_quant_g3.json

@@ -0,0 +1 @@
+{"mode": "QUANTIZE","observer": "maxabs","scale_method": "maxabs_hw","allowlist": {"types": [],"names": []},"blocklist": {"types": [],"names": []},"dump_stats_path": "/mnt/models/Llama-3.3-70B/quant/g3/inc_output"}

special_tokens_map.json

@@ -0,0 +1,4 @@
+{
+  "bos_token": "<|begin_of_text|>",
+  "eos_token": "<|eot_id|>"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920

tokenizer_config.json

@@ -0,0 +1,2064 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|finetune_right_pad_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|eom_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|python_tag|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{{- bos_token }}\n{%- if custom_tools is defined %}\n    {%- set tools = custom_tools %}\n{%- endif %}\n{%- if not tools_in_user_message is defined %}\n    {%- set tools_in_user_message = true %}\n{%- endif %}\n{%- if not date_string is defined %}\n    {%- set date_string = \"26 Jul 2024\" %}\n{%- endif %}\n{%- if not tools is defined %}\n    {%- set tools = none %}\n{%- endif %}\n\n{#- This block extracts the system message, so we can slot it into the right place. #}\n{%- if messages[0]['role'] == 'system' %}\n    {%- set system_message = messages[0]['content']|trim %}\n    {%- set messages = messages[1:] %}\n{%- else %}\n    {%- set system_message = \"\" %}\n{%- endif %}\n\n{#- System message + builtin tools #}\n{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}\n{%- if builtin_tools is defined or tools is not none %}\n    {{- \"Environment: ipython\\n\" }}\n{%- endif %}\n{%- if builtin_tools is defined %}\n    {{- \"Tools: \" + builtin_tools | reject('equalto', 'code_interpreter') | join(\", \") + \"\\n\\n\"}}\n{%- endif %}\n{{- \"Cutting Knowledge Date: December 2023\\n\" }}\n{{- \"Today Date: \" + date_string + \"\\n\\n\" }}\n{%- if tools is not none and not tools_in_user_message %}\n    {{- \"You have access to the following functions. To call a function, please respond with JSON for a function call.\" }}\n    {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n    {{- \"Do not use variables.\\n\\n\" }}\n    {%- for t in tools %}\n        {{- t | tojson(indent=4) }}\n        {{- \"\\n\\n\" }}\n    {%- endfor %}\n{%- endif %}\n{{- system_message }}\n{{- \"<|eot_id|>\" }}\n\n{#- Custom tools are passed in a user message with some extra guidance #}\n{%- if tools_in_user_message and not tools is none %}\n    {#- Extract the first user message so we can plug it in here #}\n    {%- if messages | length != 0 %}\n        {%- set first_user_message = messages[0]['content']|trim %}\n        {%- set messages = messages[1:] %}\n    {%- else %}\n        {{- raise_exception(\"Cannot put tools in the first user message when there's no first user message!\") }}\n{%- endif %}\n    {{- '<|start_header_id|>user<|end_header_id|>\\n\\n' -}}\n    {{- \"Given the following functions, please respond with a JSON for a function call \" }}\n    {{- \"with its proper arguments that best answers the given prompt.\\n\\n\" }}\n    {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n    {{- \"Do not use variables.\\n\\n\" }}\n    {%- for t in tools %}\n        {{- t | tojson(indent=4) }}\n        {{- \"\\n\\n\" }}\n    {%- endfor %}\n    {{- first_user_message + \"<|eot_id|>\"}}\n{%- endif %}\n\n{%- for message in messages %}\n    {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %}\n        {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n'+ message['content'] | trim + '<|eot_id|>' }}\n    {%- elif 'tool_calls' in message %}\n        {%- if not message.tool_calls|length == 1 %}\n            {{- raise_exception(\"This model only supports single tool-calls at once!\") }}\n        {%- endif %}\n        {%- set tool_call = message.tool_calls[0].function %}\n        {%- if builtin_tools is defined and tool_call.name in builtin_tools %}\n            {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n            {{- \"<|python_tag|>\" + tool_call.name + \".call(\" }}\n            {%- for arg_name, arg_val in tool_call.arguments | items %}\n                {{- arg_name + '=\"' + arg_val + '\"' }}\n                {%- if not loop.last %}\n                    {{- \", \" }}\n                {%- endif %}\n                {%- endfor %}\n            {{- \")\" }}\n        {%- else  %}\n            {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n            {{- '{\"name\": \"' + tool_call.name + '\", ' }}\n            {{- '\"parameters\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- \"}\" }}\n        {%- endif %}\n        {%- if builtin_tools is defined %}\n            {#- This means we're in ipython mode #}\n            {{- \"<|eom_id|>\" }}\n        {%- else %}\n            {{- \"<|eot_id|>\" }}\n        {%- endif %}\n    {%- elif message.role == \"tool\" or message.role == \"ipython\" %}\n        {{- \"<|start_header_id|>ipython<|end_header_id|>\\n\\n\" }}\n        {%- if message.content is mapping or message.content is iterable %}\n            {{- message.content | tojson }}\n        {%- else %}\n            {{- message.content }}\n        {%- endif %}\n        {{- \"<|eot_id|>\" }}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "extra_special_tokens": {},
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 131072,
+  "pad_token": "<|finetune_right_pad_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}