38d7d68a49c36a103d86e9528d836206755966b7cd408e545932061821a976ce

README.md

@@ -0,0 +1,85 @@
+---
+thumbnail: "https://assets-global.website-files.com/646b351987a8d8ce158d1940/64ec9e96b4334c0e1ac41504_Logo%20with%20white%20text.svg"
+base_model: THUDM/LongWriter-glm4-9b
+metrics:
+- memory_disk
+- memory_inference
+- inference_latency
+- inference_throughput
+- inference_CO2_emissions
+- inference_energy_consumption
+tags:
+- pruna-ai
+---
+<!-- header start -->
+<!-- 200823 -->
+<div style="width: auto; margin-left: auto; margin-right: auto">
+    <a href="https://www.pruna.ai/" target="_blank" rel="noopener noreferrer">
+        <img src="https://i.imgur.com/eDAlcgk.png" alt="PrunaAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
+    </a>
+</div>
+<!-- header end -->
+
+[![Twitter](https://img.shields.io/twitter/follow/PrunaAI?style=social)](https://twitter.com/PrunaAI)
+[![GitHub](https://img.shields.io/github/followers/PrunaAI?label=Follow%20%40PrunaAI&style=social)](https://github.com/PrunaAI)
+[![LinkedIn](https://img.shields.io/badge/LinkedIn-Connect-blue)](https://www.linkedin.com/company/93832878/admin/feed/posts/?feedType=following)
+[![Discord](https://img.shields.io/badge/Discord-Join%20Us-blue?style=social&logo=discord)](https://discord.gg/rskEr4BZJx)
+
+# Simply make AI models cheaper, smaller, faster, and greener!
+
+- Give a thumbs up if you like this model!
+- Contact us and tell us which model to compress next [here](https://www.pruna.ai/contact).
+- Request access to easily compress your *own* AI models [here](https://z0halsaff74.typeform.com/pruna-access?typeform-source=www.pruna.ai).
+- Read the documentations to know more [here](https://pruna-ai-pruna.readthedocs-hosted.com/en/latest/)
+- Join Pruna AI community on Discord [here](https://discord.gg/CP4VSgck) to share feedback/suggestions or get help.
+
+## Results
+
+![image info](./plots.png)
+
+**Frequently Asked Questions**
+- ***How does the compression work?*** The model is compressed with llm-int8.
+- ***How does the model quality change?*** The quality of the model output might vary compared to the base model.
+- ***How is the model efficiency evaluated?*** These results were obtained on HARDWARE_NAME with configuration described in `model/smash_config.json` and are obtained after a hardware warmup. The smashed model is directly compared to the original base model. Efficiency results may vary in other settings (e.g. other hardware, image size, batch size, ...). We recommend to directly run them in the use-case conditions to know if the smashed model can benefit you.
+- ***What is the model format?*** We use safetensors.
+- ***What calibration data has been used?*** If needed by the compression method, we used WikiText as the calibration data.
+- ***What is the naming convention for Pruna Huggingface models?*** We take the original model name and append "turbo", "tiny", or "green" if the smashed model has a measured inference speed, inference memory, or inference energy consumption which is less than 90% of the original base model.
+- ***How to compress my own models?*** You can request premium access to more compression methods and tech support for your specific use-cases [here](https://z0halsaff74.typeform.com/pruna-access?typeform-source=www.pruna.ai).
+- ***What are "first" metrics?*** Results mentioning "first" are obtained after the first run of the model. The first run might take more memory or be slower than the subsequent runs due cuda overheads.
+- ***What are "Sync" and "Async" metrics?*** "Sync" metrics are obtained by syncing all GPU processes and stop measurement when all of them are executed. "Async" metrics are obtained without syncing all GPU processes and stop when the model output can be used by the CPU. We provide both metrics since both could be relevant depending on the use-case. We recommend to test the efficiency gains directly in your use-cases.
+
+## Setup
+
+You can run the smashed model with these steps:
+
+0. Check requirements from the original repo THUDM/LongWriter-glm4-9b installed. In particular, check python, cuda, and transformers versions.
+1. Make sure that you have installed quantization related packages.
+    ```bash
+    pip install transformers accelerate bitsandbytes>0.37.0
+    ```
+2. Load & run the model.
+    ```python 
+   from transformers import AutoModelForCausalLM, AutoTokenizer
+   
+
+   model = AutoModelForCausalLM.from_pretrained("PrunaAI/THUDM-LongWriter-glm4-9b-bnb-4bit-smashed", trust_remote_code=True, device_map='auto')
+   tokenizer = AutoTokenizer.from_pretrained("THUDM/LongWriter-glm4-9b")
+    
+   input_ids = tokenizer("What is the color of prunes?,", return_tensors='pt').to(model.device)["input_ids"]
+    
+   outputs = model.generate(input_ids, max_new_tokens=216)
+   tokenizer.decode(outputs[0])
+    ```
+
+## Configurations
+
+The configuration info are in `smash_config.json`.
+
+## Credits & License
+
+The license of the smashed model follows the license of the original model. Please check the license of the original model THUDM/LongWriter-glm4-9b before using this model which provided the base model. The license  of the `pruna-engine` is [here](https://pypi.org/project/pruna-engine/) on Pypi.
+
+## Want to compress other models?
+
+- Contact us and tell us which model to compress next [here](https://www.pruna.ai/contact).
+- Request access to easily compress your own AI models [here](https://z0halsaff74.typeform.com/pruna-access?typeform-source=www.pruna.ai).

config.json

@@ -0,0 +1,67 @@
+{
+  "_name_or_path": "/ceph/hdd/staff/charpent/.cache/modelsdhkyhu_nhop2bfo6",
+  "add_bias_linear": false,
+  "add_qkv_bias": true,
+  "apply_query_key_layer_scaling": true,
+  "apply_residual_connection_post_layernorm": false,
+  "architectures": [
+    "ChatGLMForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "attention_softmax_in_fp32": true,
+  "auto_map": {
+    "AutoConfig": "configuration_chatglm.ChatGLMConfig",
+    "AutoModel": "THUDM/LongWriter-glm4-9b--modeling_chatglm.ChatGLMForConditionalGeneration",
+    "AutoModelForCausalLM": "modeling_chatglm.ChatGLMForConditionalGeneration",
+    "AutoModelForSeq2SeqLM": "THUDM/LongWriter-glm4-9b--modeling_chatglm.ChatGLMForConditionalGeneration",
+    "AutoModelForSequenceClassification": "THUDM/LongWriter-glm4-9b--modeling_chatglm.ChatGLMForSequenceClassification"
+  },
+  "bias_dropout_fusion": true,
+  "classifier_dropout": null,
+  "eos_token_id": [
+    151329,
+    151336,
+    151338
+  ],
+  "ffn_hidden_size": 13696,
+  "fp32_residual_connection": false,
+  "hidden_dropout": 0.0,
+  "hidden_size": 4096,
+  "kv_channels": 128,
+  "layernorm_epsilon": 1.5625e-07,
+  "model_type": "chatglm",
+  "multi_query_attention": true,
+  "multi_query_group_num": 2,
+  "num_attention_heads": 32,
+  "num_hidden_layers": 40,
+  "num_layers": 40,
+  "original_rope": true,
+  "pad_token_id": 151329,
+  "padded_vocab_size": 151552,
+  "post_layer_norm": true,
+  "quantization_config": {
+    "_load_in_4bit": true,
+    "_load_in_8bit": false,
+    "bnb_4bit_compute_dtype": "bfloat16",
+    "bnb_4bit_quant_storage": "uint8",
+    "bnb_4bit_quant_type": "fp4",
+    "bnb_4bit_use_double_quant": false,
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_has_fp16_weight": false,
+    "llm_int8_skip_modules": [
+      "lm_head"
+    ],
+    "llm_int8_threshold": 6.0,
+    "load_in_4bit": true,
+    "load_in_8bit": false,
+    "quant_method": "bitsandbytes"
+  },
+  "rmsnorm": true,
+  "rope_ratio": 500,
+  "seq_length": 1048576,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.42.4",
+  "use_cache": true,
+  "vocab_size": 151552
+}

configuration_chatglm.py

@@ -0,0 +1,58 @@
+from transformers import PretrainedConfig
+
+
+class ChatGLMConfig(PretrainedConfig):
+    model_type = "chatglm"
+
+    def __init__(
+            self,
+            num_layers=28,
+            padded_vocab_size=65024,
+            hidden_size=4096,
+            ffn_hidden_size=13696,
+            kv_channels=128,
+            num_attention_heads=32,
+            seq_length=2048,
+            hidden_dropout=0.0,
+            classifier_dropout=None,
+            attention_dropout=0.0,
+            layernorm_epsilon=1e-5,
+            rmsnorm=True,
+            apply_residual_connection_post_layernorm=False,
+            post_layer_norm=True,
+            add_bias_linear=False,
+            add_qkv_bias=False,
+            bias_dropout_fusion=True,
+            multi_query_attention=False,
+            multi_query_group_num=1,
+            rope_ratio=1,
+            apply_query_key_layer_scaling=True,
+            attention_softmax_in_fp32=True,
+            fp32_residual_connection=False,
+            **kwargs
+    ):
+        self.num_layers = num_layers
+        self.vocab_size = padded_vocab_size
+        self.padded_vocab_size = padded_vocab_size
+        self.hidden_size = hidden_size
+        self.ffn_hidden_size = ffn_hidden_size
+        self.kv_channels = kv_channels
+        self.num_attention_heads = num_attention_heads
+        self.seq_length = seq_length
+        self.hidden_dropout = hidden_dropout
+        self.classifier_dropout = classifier_dropout
+        self.attention_dropout = attention_dropout
+        self.layernorm_epsilon = layernorm_epsilon
+        self.rmsnorm = rmsnorm
+        self.apply_residual_connection_post_layernorm = apply_residual_connection_post_layernorm
+        self.post_layer_norm = post_layer_norm
+        self.add_bias_linear = add_bias_linear
+        self.add_qkv_bias = add_qkv_bias
+        self.bias_dropout_fusion = bias_dropout_fusion
+        self.multi_query_attention = multi_query_attention
+        self.multi_query_group_num = multi_query_group_num
+        self.rope_ratio = rope_ratio
+        self.apply_query_key_layer_scaling = apply_query_key_layer_scaling
+        self.attention_softmax_in_fp32 = attention_softmax_in_fp32
+        self.fp32_residual_connection = fp32_residual_connection
+        super().__init__(**kwargs)

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "_from_model_config": true,
+  "eos_token_id": [
+    151329,
+    151336,
+    151338
+  ],
+  "pad_token_id": 151329,
+  "transformers_version": "4.42.4"
+}

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b27bdda7afa891460dde7825493ceb8bddef2412a7e55213fbebe8b95856c24
+size 1183988127

model.safetensors.index.json

@@ -0,0 +1,774 @@
+{
+  "metadata": {
+    "total_size": 6180576195
+  },
+  "weight_map": {
+    "transformer.embedding.word_embeddings.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.final_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.0.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.1.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.10.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.11.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.12.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.13.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.14.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.15.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.16.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.17.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.18.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.19.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.2.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.20.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.21.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.22.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.23.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.24.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.25.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.26.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.27.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.28.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.29.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.3.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.30.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.31.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.32.mlp.dense_4h_to_h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.32.mlp.dense_4h_to_h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.32.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.32.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.32.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.33.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.mlp.dense_4h_to_h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.mlp.dense_4h_to_h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.mlp.dense_h_to_4h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.mlp.dense_h_to_4h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.self_attention.dense.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.self_attention.dense.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.self_attention.dense.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.self_attention.query_key_value.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.self_attention.query_key_value.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.33.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.mlp.dense_4h_to_h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.mlp.dense_4h_to_h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.mlp.dense_h_to_4h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.mlp.dense_h_to_4h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.self_attention.dense.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.self_attention.dense.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.self_attention.dense.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.self_attention.query_key_value.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.self_attention.query_key_value.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.34.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.mlp.dense_4h_to_h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.mlp.dense_4h_to_h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.mlp.dense_h_to_4h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.mlp.dense_h_to_4h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.self_attention.dense.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.self_attention.dense.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.self_attention.dense.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.self_attention.query_key_value.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.self_attention.query_key_value.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.35.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.mlp.dense_4h_to_h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.mlp.dense_4h_to_h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.mlp.dense_h_to_4h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.mlp.dense_h_to_4h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.self_attention.dense.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.self_attention.dense.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.self_attention.dense.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.self_attention.query_key_value.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.self_attention.query_key_value.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.36.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.mlp.dense_4h_to_h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.mlp.dense_4h_to_h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.mlp.dense_h_to_4h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.mlp.dense_h_to_4h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.self_attention.dense.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.self_attention.dense.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.self_attention.dense.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.self_attention.query_key_value.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.self_attention.query_key_value.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.37.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.mlp.dense_4h_to_h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.mlp.dense_4h_to_h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.mlp.dense_h_to_4h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.mlp.dense_h_to_4h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.self_attention.dense.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.self_attention.dense.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.self_attention.dense.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.self_attention.query_key_value.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.self_attention.query_key_value.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.38.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.mlp.dense_4h_to_h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.mlp.dense_4h_to_h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.mlp.dense_h_to_4h.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.mlp.dense_h_to_4h.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.self_attention.dense.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.self_attention.dense.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.self_attention.dense.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.self_attention.query_key_value.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.self_attention.query_key_value.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.39.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.encoder.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.4.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.5.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.6.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.7.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.8.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.mlp.dense_4h_to_h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.mlp.dense_4h_to_h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.mlp.dense_4h_to_h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.mlp.dense_h_to_4h.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.mlp.dense_h_to_4h.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.mlp.dense_h_to_4h.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.self_attention.dense.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.self_attention.dense.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.self_attention.dense.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.self_attention.dense.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.self_attention.query_key_value.weight.absmax": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.self_attention.query_key_value.weight.quant_map": "model-00001-of-00002.safetensors",
+    "transformer.encoder.layers.9.self_attention.query_key_value.weight.quant_state.bitsandbytes__fp4": "model-00001-of-00002.safetensors",
+    "transformer.output_layer.weight": "model-00002-of-00002.safetensors",
+    "transformer.output_layer.weight.absmax": "model-00002-of-00002.safetensors",
+    "transformer.output_layer.weight.quant_map": "model-00002-of-00002.safetensors",
+    "transformer.output_layer.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00002.safetensors",
+    "transformer.rotary_pos_emb.inv_freq": "model-00001-of-00002.safetensors"
+  }
+}

modeling_chatglm.py

@@ -0,0 +1,898 @@
+""" PyTorch ChatGLM model. """
+
+import math
+import copy
+import warnings
+import re
+import sys
+
+import torch
+import torch.utils.checkpoint
+import torch.nn.functional as F
+from torch import nn
+from torch.nn import CrossEntropyLoss, LayerNorm
+from torch.nn.utils import skip_init
+from typing import Optional, Tuple, Union, List, Callable, Dict, Any
+
+from transformers.modeling_outputs import (
+    BaseModelOutputWithPast,
+    CausalLMOutputWithPast,
+)
+from transformers.modeling_utils import PreTrainedModel
+from transformers.utils import logging
+from transformers.generation.logits_process import LogitsProcessor
+from transformers.generation.utils import LogitsProcessorList, StoppingCriteriaList, GenerationConfig, ModelOutput
+
+from .configuration_chatglm import ChatGLMConfig
+from einops import rearrange
+try:
+    from flash_attn.flash_attn_interface import flash_attn_unpadded_func
+except ImportError:
+    try:
+        # FlashAttention-2
+        from flash_attn.flash_attn_interface import flash_attn_varlen_func as flash_attn_unpadded_func
+    except ImportError:
+        flash_attn_unpadded_func = None
+
+# flags required to enable jit fusion kernels
+
+if sys.platform != 'darwin':
+    torch._C._jit_set_profiling_mode(False)
+    torch._C._jit_set_profiling_executor(False)
+    torch._C._jit_override_can_fuse_on_cpu(True)
+    torch._C._jit_override_can_fuse_on_gpu(True)
+
+logger = logging.get_logger(__name__)
+
+_CHECKPOINT_FOR_DOC = "THUDM/ChatGLM2-6B"
+_CONFIG_FOR_DOC = "ChatGLM6BConfig"
+
+CHATGLM_6B_PRETRAINED_MODEL_ARCHIVE_LIST = [
+    "THUDM/chatglm2-6b",
+    # See all ChatGLM models at https://huggingface.co/models?filter=chatglm
+]
+
+def default_init(cls, *args, **kwargs):
+    return cls(*args, **kwargs)
+
+
+class InvalidScoreLogitsProcessor(LogitsProcessor):
+    def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTensor) -> torch.FloatTensor:
+        if torch.isnan(scores).any() or torch.isinf(scores).any():
+            scores.zero_()
+            scores[..., 5] = 5e4
+        return scores
+
+def split_tensor_along_last_dim(
+        tensor: torch.Tensor,
+        num_partitions: int,
+        contiguous_split_chunks: bool = False,
+) -> List[torch.Tensor]:
+    """Split a tensor along its last dimension.
+
+    Arguments:
+        tensor: input tensor.
+        num_partitions: number of partitions to split the tensor
+        contiguous_split_chunks: If True, make each chunk contiguous
+                                 in memory.
+
+    Returns:
+        A list of Tensors
+    """
+    # Get the size and dimension.
+    last_dim = tensor.dim() - 1
+    last_dim_size = tensor.size()[last_dim] // num_partitions
+    # Split.
+    tensor_list = torch.split(tensor, last_dim_size, dim=last_dim)
+    # Note: torch.split does not create contiguous tensors by default.
+    if contiguous_split_chunks:
+        return tuple(chunk.contiguous() for chunk in tensor_list)
+
+    return tensor_list
+
+
+class RotaryEmbedding(nn.Module):
+    def __init__(self, dim, rope_ratio=1, original_impl=False, device=None, dtype=None):
+        super().__init__()
+        inv_freq = 1.0 / (10000 ** (torch.arange(0, dim, 2, device=device).to(dtype=dtype) / dim))
+        self.register_buffer("inv_freq", inv_freq)
+        self.dim = dim
+        self.original_impl = original_impl
+        self.rope_ratio = rope_ratio
+
+    def forward_impl(
+            self, seq_len: int, n_elem: int, dtype: torch.dtype, device: torch.device, base: int = 10000
+    ):
+        """Enhanced Transformer with Rotary Position Embedding.
+
+        Derived from: https://github.com/labmlai/annotated_deep_learning_paper_implementations/blob/master/labml_nn/
+        transformers/rope/__init__.py. MIT License:
+        https://github.com/labmlai/annotated_deep_learning_paper_implementations/blob/master/license.
+        """
+        # $\Theta = {\theta_i = 10000^{\frac{2(i-1)}{d}}, i \in [1, 2, ..., \frac{d}{2}]}$
+        
+        base = base * self.rope_ratio
+        theta = 1.0 / (base ** (torch.arange(0, n_elem, 2, dtype=torch.float, device=device) / n_elem))
+
+        # Create position indexes `[0, 1, ..., seq_len - 1]`
+        seq_idx = torch.arange(seq_len, dtype=torch.float, device=device)
+
+        # Calculate the product of position index and $\theta_i$
+        idx_theta = torch.outer(seq_idx, theta).float()
+
+        cache = torch.stack([torch.cos(idx_theta), torch.sin(idx_theta)], dim=-1)
+
+        # this is to mimic the behaviour of complex32, else we will get different results
+        if dtype in (torch.float16, torch.bfloat16, torch.int8):
+            cache = cache.bfloat16() if dtype == torch.bfloat16 else cache.half()
+        return cache
+
+    def forward(self, max_seq_len, offset=0):
+        return self.forward_impl(
+            max_seq_len, self.dim, dtype=self.inv_freq.dtype, device=self.inv_freq.device
+        )
+
+
+@torch.jit.script
+def apply_rotary_pos_emb(x: torch.Tensor, rope_cache: torch.Tensor) -> torch.Tensor:
+    # x: [sq, b, np, hn]
+    sq, b, np, hn = x.size(0), x.size(1), x.size(2), x.size(3)
+    rot_dim = rope_cache.shape[-2] * 2
+    x, x_pass = x[..., :rot_dim], x[..., rot_dim:]
+    # truncate to support variable sizes
+    rope_cache = rope_cache[:sq]
+    xshaped = x.reshape(sq, -1, np, rot_dim // 2, 2)
+    rope_cache = rope_cache.view(sq, -1, 1, xshaped.size(3), 2)
+    x_out2 = torch.stack(
+        [
+            xshaped[..., 0] * rope_cache[..., 0] - xshaped[..., 1] * rope_cache[..., 1],
+            xshaped[..., 1] * rope_cache[..., 0] + xshaped[..., 0] * rope_cache[..., 1],
+        ],
+        -1,
+    )
+    x_out2 = x_out2.flatten(3)
+    return torch.cat((x_out2, x_pass), dim=-1)
+
+
+class RMSNorm(torch.nn.Module):
+    def __init__(self, normalized_shape, eps=1e-5, device=None, dtype=None, **kwargs):
+        super().__init__()
+        self.weight = torch.nn.Parameter(torch.empty(normalized_shape, device=device, dtype=dtype))
+        self.eps = eps
+
+    def forward(self, hidden_states: torch.Tensor):
+        input_dtype = hidden_states.dtype
+        variance = hidden_states.to(torch.float32).pow(2).mean(-1, keepdim=True)
+        hidden_states = hidden_states * torch.rsqrt(variance + self.eps)
+
+        return (self.weight * hidden_states).to(input_dtype)
+
+
+class CoreAttention(torch.nn.Module):
+    def __init__(self, config: ChatGLMConfig, layer_number):
+        super(CoreAttention, self).__init__()
+
+        self.apply_query_key_layer_scaling = config.apply_query_key_layer_scaling
+        self.attention_softmax_in_fp32 = config.attention_softmax_in_fp32
+        if self.apply_query_key_layer_scaling:
+            self.attention_softmax_in_fp32 = True
+        self.layer_number = max(1, layer_number)
+
+        projection_size = config.kv_channels * config.num_attention_heads
+
+        # Per attention head and per partition values.
+        self.hidden_size_per_partition = projection_size
+        self.hidden_size_per_attention_head = projection_size // config.num_attention_heads
+        self.num_attention_heads_per_partition = config.num_attention_heads
+
+        self.norm_factor = math.sqrt(self.hidden_size_per_attention_head)
+        self.attention_dropout = config.attention_dropout
+
+    def forward(self, query_layer, key_layer, value_layer, attention_mask):
+        seqlen_q, batch_size = query_layer.shape[0], query_layer.shape[1]
+        seqlen_k = key_layer.shape[0]
+        query_layer, key_layer, value_layer = [rearrange(x, 's b ... -> (b s) ...') for x in [query_layer, key_layer, value_layer]]
+        # DO flash_attn_varlen_func
+        if attention_mask is None or attention_mask.ndim != 1:
+            cu_seqlens_q = torch.arange(0, (batch_size + 1) * seqlen_q, step=seqlen_q, dtype=torch.int32,
+                                    device=query_layer.device)
+        else:
+            assert seqlen_q == seqlen_k
+            cu_seqlens_q = attention_mask
+        if self.training:
+            assert seqlen_k == seqlen_q
+            is_causal = True
+            cu_seqlens_k = cu_seqlens_q
+        else:
+            is_causal = seqlen_q == seqlen_k
+            cu_seqlens_k = torch.arange(0, (batch_size + 1) * seqlen_k, step=seqlen_k, dtype=torch.int32,
+                        device=query_layer.device) if not is_causal else cu_seqlens_q
+            self.attention_dropout = 0
+        context_layer = flash_attn_unpadded_func(
+            query_layer, key_layer, value_layer, cu_seqlens_q, cu_seqlens_k, seqlen_q, seqlen_k,
+            self.attention_dropout,
+            softmax_scale=1.0 / self.norm_factor, causal=is_causal
+        )
+        context_layer = rearrange(context_layer, '(b s) ... -> s b ...', b=batch_size)
+        new_context_layer_shape = context_layer.size()[:-2] + (self.hidden_size_per_partition,)
+        context_layer = context_layer.reshape(*new_context_layer_shape)
+        return context_layer
+
+
+class SelfAttention(torch.nn.Module):
+    """Parallel self-attention layer abstract class.
+
+    Self-attention layer takes input with size [s, b, h]
+    and returns output of the same size.
+    """
+
+    def __init__(self, config: ChatGLMConfig, layer_number, device=None):
+        super(SelfAttention, self).__init__()
+        self.layer_number = max(1, layer_number)
+
+        self.projection_size = config.kv_channels * config.num_attention_heads
+
+        # Per attention head and per partition values.
+        self.hidden_size_per_attention_head = self.projection_size // config.num_attention_heads
+        self.num_attention_heads_per_partition = config.num_attention_heads
+
+        self.multi_query_attention = config.multi_query_attention
+        self.qkv_hidden_size = 3 * self.projection_size
+        if self.multi_query_attention:
+            self.num_multi_query_groups_per_partition = config.multi_query_group_num
+            self.qkv_hidden_size = (
+                    self.projection_size + 2 * self.hidden_size_per_attention_head * config.multi_query_group_num
+            )
+        self.query_key_value = nn.Linear(config.hidden_size, self.qkv_hidden_size,
+                                         bias=config.add_bias_linear or config.add_qkv_bias,
+                                         device=device, **_config_to_kwargs(config)
+                                         )
+
+        self.core_attention = CoreAttention(config, self.layer_number)
+
+        # Output.
+        self.dense = nn.Linear(self.projection_size, config.hidden_size, bias=config.add_bias_linear,
+                               device=device, **_config_to_kwargs(config)
+                               )
+
+    def _allocate_memory(self, inference_max_sequence_len, batch_size, device=None, dtype=None):
+        if self.multi_query_attention:
+            num_attention_heads = self.num_multi_query_groups_per_partition
+        else:
+            num_attention_heads = self.num_attention_heads_per_partition
+        return torch.empty(
+            inference_max_sequence_len,
+            batch_size,
+            num_attention_heads,
+            self.hidden_size_per_attention_head,
+            dtype=dtype,
+            device=device,
+        )
+
+    def forward(
+            self, hidden_states, attention_mask, rotary_pos_emb, kv_cache=None, use_cache=True
+    ):
+        # hidden_states: [sq, b, h]
+
+        # =================================================
+        # Pre-allocate memory for key-values for inference.
+        # =================================================
+        # =====================
+        # Query, Key, and Value
+        # =====================
+
+        # Attention heads [sq, b, h] --> [sq, b, (np * 3 * hn)]
+        mixed_x_layer = self.query_key_value(hidden_states)
+
+        if self.multi_query_attention:
+            (query_layer, key_layer, value_layer) = mixed_x_layer.split(
+                [
+                    self.num_attention_heads_per_partition * self.hidden_size_per_attention_head,
+                    self.num_multi_query_groups_per_partition * self.hidden_size_per_attention_head,
+                    self.num_multi_query_groups_per_partition * self.hidden_size_per_attention_head,
+                ],
+                dim=-1,
+            )
+            query_layer = query_layer.view(
+                query_layer.size()[:-1] + (self.num_attention_heads_per_partition, self.hidden_size_per_attention_head)
+            )
+            key_layer = key_layer.view(
+                key_layer.size()[:-1] + (self.num_multi_query_groups_per_partition, self.hidden_size_per_attention_head)
+            )
+            value_layer = value_layer.view(
+                value_layer.size()[:-1]
+                + (self.num_multi_query_groups_per_partition, self.hidden_size_per_attention_head)
+            )
+        else:
+            new_tensor_shape = mixed_x_layer.size()[:-1] + \
+                               (self.num_attention_heads_per_partition,
+                                3 * self.hidden_size_per_attention_head)
+            mixed_x_layer = mixed_x_layer.view(*new_tensor_shape)
+
+            # [sq, b, np, 3 * hn] --> 3 [sq, b, np, hn]
+            (query_layer, key_layer, value_layer) = split_tensor_along_last_dim(mixed_x_layer, 3)
+
+        # apply relative positional encoding (rotary embedding)
+        if rotary_pos_emb is not None:
+            query_layer = apply_rotary_pos_emb(query_layer, rotary_pos_emb)
+            key_layer = apply_rotary_pos_emb(key_layer, rotary_pos_emb)
+
+        # adjust key and value for inference
+        if use_cache:
+            if kv_cache is not None:
+                cache_k, cache_v = kv_cache
+                key_layer = torch.cat((cache_k, key_layer), dim=0)
+                value_layer = torch.cat((cache_v, value_layer), dim=0)
+            kv_cache = (key_layer, value_layer)
+        else:
+            kv_cache = None
+        
+            
+        if self.multi_query_attention:
+            key_layer = key_layer.unsqueeze(-2)
+            key_layer = key_layer.expand(
+                -1, -1, -1, self.num_attention_heads_per_partition // self.num_multi_query_groups_per_partition, -1
+            )
+            key_layer = key_layer.contiguous().view(
+                key_layer.size()[:2] + (self.num_attention_heads_per_partition, self.hidden_size_per_attention_head)
+            )
+            value_layer = value_layer.unsqueeze(-2)
+            value_layer = value_layer.expand(
+                -1, -1, -1, self.num_attention_heads_per_partition // self.num_multi_query_groups_per_partition, -1
+            )
+            value_layer = value_layer.contiguous().view(
+                value_layer.size()[:2] + (self.num_attention_heads_per_partition, self.hidden_size_per_attention_head)
+            )
+
+        # ==================================
+        # core attention computation
+        # ==================================
+
+        context_layer = self.core_attention(query_layer, key_layer, value_layer, attention_mask)
+
+        # =================
+        # Output. [sq, b, h]
+        # =================
+
+        output = self.dense(context_layer)
+
+        return output, kv_cache
+
+
+def _config_to_kwargs(args):
+    common_kwargs = {
+        "dtype": args.torch_dtype,
+    }
+    return common_kwargs
+
+
+class MLP(torch.nn.Module):
+    """MLP.
+
+    MLP will take the input with h hidden state, project it to 4*h
+    hidden dimension, perform nonlinear transformation, and project the
+    state back into h hidden dimension.
+    """
+
+    def __init__(self, config: ChatGLMConfig, device=None):
+        super(MLP, self).__init__()
+
+        self.add_bias = config.add_bias_linear
+
+        # Project to 4h. If using swiglu double the output width, see https://arxiv.org/pdf/2002.05202.pdf
+        self.dense_h_to_4h = nn.Linear(
+            config.hidden_size,
+            config.ffn_hidden_size * 2,
+            bias=self.add_bias,
+            device=device,
+            **_config_to_kwargs(config)
+        )
+
+        def swiglu(x):
+            x = torch.chunk(x, 2, dim=-1)
+            return F.silu(x[0]) * x[1]
+
+        self.activation_func = swiglu
+
+        # Project back to h.
+        self.dense_4h_to_h = nn.Linear(
+            config.ffn_hidden_size,
+            config.hidden_size,
+            bias=self.add_bias,
+            device=device,
+            **_config_to_kwargs(config)
+        )
+
+    def forward(self, hidden_states):
+        # [s, b, 4hp]
+        intermediate_parallel = self.dense_h_to_4h(hidden_states)
+        intermediate_parallel = self.activation_func(intermediate_parallel)
+        # [s, b, h]
+        output = self.dense_4h_to_h(intermediate_parallel)
+        return output
+
+
+class GLMBlock(torch.nn.Module):
+    """A single transformer layer.
+
+    Transformer layer takes input with size [s, b, h] and returns an
+    output of the same size.
+    """
+
+    def __init__(self, config: ChatGLMConfig, layer_number, device=None):
+        super(GLMBlock, self).__init__()
+        self.layer_number = layer_number
+
+        self.apply_residual_connection_post_layernorm = config.apply_residual_connection_post_layernorm
+
+        self.fp32_residual_connection = config.fp32_residual_connection
+
+        LayerNormFunc = RMSNorm if config.rmsnorm else LayerNorm
+        # Layernorm on the input data.
+        self.input_layernorm = LayerNormFunc(config.hidden_size, eps=config.layernorm_epsilon, device=device,
+                                             dtype=config.torch_dtype)
+
+        # Self attention.
+        self.self_attention = SelfAttention(config, layer_number, device=device)
+        self.hidden_dropout = config.hidden_dropout
+
+        # Layernorm on the attention output
+        self.post_attention_layernorm = LayerNormFunc(config.hidden_size, eps=config.layernorm_epsilon, device=device,
+                                                      dtype=config.torch_dtype)
+
+        # MLP
+        self.mlp = MLP(config, device=device)
+
+    def forward(
+            self, hidden_states, attention_mask, rotary_pos_emb, kv_cache=None, use_cache=True,
+    ):
+        # hidden_states: [s, b, h]
+
+        # Layer norm at the beginning of the transformer layer.
+        layernorm_output = self.input_layernorm(hidden_states)
+        # Self attention.
+        attention_output, kv_cache = self.self_attention(
+            layernorm_output,
+            attention_mask,
+            rotary_pos_emb,
+            kv_cache=kv_cache,
+            use_cache=use_cache
+        )
+
+        # Residual connection.
+        if self.apply_residual_connection_post_layernorm:
+            residual = layernorm_output
+        else:
+            residual = hidden_states
+
+        layernorm_input = torch.nn.functional.dropout(attention_output, p=self.hidden_dropout, training=self.training)
+        layernorm_input = residual + layernorm_input
+
+        # Layer norm post the self attention.
+        layernorm_output = self.post_attention_layernorm(layernorm_input)
+
+        # MLP.
+        mlp_output = self.mlp(layernorm_output)
+
+        # Second residual connection.
+        if self.apply_residual_connection_post_layernorm:
+            residual = layernorm_output
+        else:
+            residual = layernorm_input
+
+        output = torch.nn.functional.dropout(mlp_output, p=self.hidden_dropout, training=self.training)
+        output = residual + output
+
+        return output, kv_cache
+
+
+class GLMTransformer(torch.nn.Module):
+    """Transformer class."""
+
+    def __init__(self, config: ChatGLMConfig, device=None):
+        super(GLMTransformer, self).__init__()
+
+        self.fp32_residual_connection = config.fp32_residual_connection
+        self.post_layer_norm = config.post_layer_norm
+
+        # Number of layers.
+        self.num_layers = config.num_layers
+
+        # Transformer layers.
+        def build_layer(layer_number):
+            return GLMBlock(config, layer_number, device=device)
+
+        self.layers = torch.nn.ModuleList([build_layer(i + 1) for i in range(self.num_layers)])
+
+        if self.post_layer_norm:
+            LayerNormFunc = RMSNorm if config.rmsnorm else LayerNorm
+            # Final layer norm before output.
+            self.final_layernorm = LayerNormFunc(config.hidden_size, eps=config.layernorm_epsilon, device=device,
+                                                 dtype=config.torch_dtype)
+
+        self.gradient_checkpointing = False
+
+    def _get_layer(self, layer_number):
+        return self.layers[layer_number]
+
+    def forward(
+            self, hidden_states, attention_mask, rotary_pos_emb, kv_caches=None,
+            use_cache: Optional[bool] = True,
+            output_hidden_states: Optional[bool] = False,
+    ):
+        if not kv_caches:
+            kv_caches = [None for _ in range(self.num_layers)]
+        presents = () if use_cache else None
+        if self.gradient_checkpointing and self.training:
+            if use_cache:
+                # logger.warning_once(
+                #     "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..."
+                # )
+                use_cache = False
+
+        all_self_attentions = None
+        all_hidden_states = () if output_hidden_states else None
+        for index in range(self.num_layers):
+            if output_hidden_states:
+                all_hidden_states = all_hidden_states + (hidden_states,)
+
+            layer = self._get_layer(index)
+            if self.gradient_checkpointing and self.training:
+                layer_ret = torch.utils.checkpoint.checkpoint(
+                    layer,
+                    hidden_states,
+                    attention_mask,
+                    rotary_pos_emb,
+                    kv_caches[index],
+                    use_cache,
+                    use_reentrant=False
+                )
+            else:
+                layer_ret = layer(
+                    hidden_states,
+                    attention_mask,
+                    rotary_pos_emb,
+                    kv_cache=kv_caches[index],
+                    use_cache=use_cache
+                )
+            hidden_states, kv_cache = layer_ret
+            if use_cache:
+                presents = presents + (kv_cache,)
+
+        if output_hidden_states:
+            all_hidden_states = all_hidden_states + (hidden_states,)
+
+        # Final layer norm.
+        if self.post_layer_norm:
+            hidden_states = self.final_layernorm(hidden_states)
+
+        return hidden_states, presents, all_hidden_states, all_self_attentions
+
+
+class ChatGLMPreTrainedModel(PreTrainedModel):
+    """
+    An abstract class to handle weights initialization and
+    a simple interface for downloading and loading pretrained models.
+    """
+
+    is_parallelizable = False
+    supports_gradient_checkpointing = True
+    config_class = ChatGLMConfig
+    base_model_prefix = "transformer"
+    _no_split_modules = ["GLMBlock"]
+
+    def _init_weights(self, module: nn.Module):
+        """Initialize the weights."""
+        return
+
+    def get_masks(self, input_ids, past_key_values, padding_mask=None):
+        batch_size, seq_length = input_ids.shape
+        full_attention_mask = torch.ones(batch_size, seq_length, seq_length, device=input_ids.device)
+        full_attention_mask.tril_()
+        past_length = 0
+        if past_key_values:
+            past_length = past_key_values[0][0].shape[0]
+        if past_length:
+            full_attention_mask = torch.cat((torch.ones(batch_size, seq_length, past_length,
+                                                        device=input_ids.device), full_attention_mask), dim=-1)
+        if padding_mask is not None:
+            full_attention_mask = full_attention_mask * padding_mask.unsqueeze(1)
+        if not past_length and padding_mask is not None:
+            full_attention_mask -= padding_mask.unsqueeze(-1) - 1
+        full_attention_mask = (full_attention_mask < 0.5).bool()
+        full_attention_mask.unsqueeze_(1)
+        return full_attention_mask
+
+    def get_position_ids(self, input_ids, device):
+        batch_size, seq_length = input_ids.shape
+        position_ids = torch.arange(seq_length, dtype=torch.long, device=device).unsqueeze(0).repeat(batch_size, 1)
+        return position_ids
+
+    def _set_gradient_checkpointing(self, module, value=False):
+        if isinstance(module, GLMTransformer):
+            module.gradient_checkpointing = value
+
+
+class Embedding(torch.nn.Module):
+    """Language model embeddings."""
+
+    def __init__(self, config: ChatGLMConfig, device=None):
+        super(Embedding, self).__init__()
+
+        self.hidden_size = config.hidden_size
+        # Word embeddings (parallel).
+        self.word_embeddings = nn.Embedding(
+            config.padded_vocab_size,
+            self.hidden_size,
+            dtype=config.torch_dtype,
+            device=device
+        )
+        self.fp32_residual_connection = config.fp32_residual_connection
+
+    def forward(self, input_ids):
+        # Embeddings.
+        words_embeddings = self.word_embeddings(input_ids)
+        embeddings = words_embeddings
+        # Data format change to avoid explicit tranposes : [b s h] --> [s b h].
+        embeddings = embeddings.transpose(0, 1).contiguous()
+        # If the input flag for fp32 residual connection is set, convert for float.
+        if self.fp32_residual_connection:
+            embeddings = embeddings.float()
+        return embeddings
+
+
+class ChatGLMModel(ChatGLMPreTrainedModel):
+    def __init__(self, config: ChatGLMConfig, device=None, empty_init=True):
+        super().__init__(config)
+        if empty_init:
+            init_method = skip_init
+        else:
+            init_method = default_init
+        init_kwargs = {}
+        if device is not None:
+            init_kwargs["device"] = device
+        self.embedding = init_method(Embedding, config, **init_kwargs)
+
+        # Rotary positional embeddings
+        self.seq_length = config.seq_length
+        rotary_dim = (
+            config.hidden_size // config.num_attention_heads if config.kv_channels is None else config.kv_channels
+        )
+
+        self.rotary_pos_emb = RotaryEmbedding(rotary_dim // 2, rope_ratio=config.rope_ratio, original_impl=config.original_rope, 
+                                              device=device, dtype=config.torch_dtype)
+        self.encoder = init_method(GLMTransformer, config, **init_kwargs)
+        self.output_layer = init_method(nn.Linear, config.hidden_size, config.padded_vocab_size, bias=False,
+                                        dtype=config.torch_dtype, **init_kwargs)
+
+    def get_input_embeddings(self):
+        return self.embedding.word_embeddings
+
+    def forward(
+            self,
+            input_ids,
+            position_ids: Optional[torch.Tensor] = None,
+            attention_mask: Optional[torch.BoolTensor] = None,
+            full_attention_mask: Optional[torch.BoolTensor] = None,
+            past_key_values: Optional[Tuple[Tuple[torch.Tensor, torch.Tensor], ...]] = None,
+            inputs_embeds: Optional[torch.Tensor] = None,
+            use_cache: Optional[bool] = None,
+            output_hidden_states: Optional[bool] = None,
+            return_dict: Optional[bool] = None,
+    ):
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        use_cache = use_cache if use_cache is not None else self.config.use_cache
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        batch_size, seq_length = input_ids.shape
+
+        if inputs_embeds is None:
+            inputs_embeds = self.embedding(input_ids)
+
+        # if full_attention_mask is None:
+        #     if (attention_mask is not None and not attention_mask.all()) or (past_key_values and seq_length != 1):
+        #         full_attention_mask = self.get_masks(input_ids, past_key_values, padding_mask=attention_mask)
+
+        # Rotary positional embeddings
+        rotary_pos_emb = self.rotary_pos_emb(self.seq_length)
+        if position_ids is not None:
+            rotary_pos_emb = rotary_pos_emb[position_ids]
+        else:
+            rotary_pos_emb = rotary_pos_emb[None, :seq_length]
+        rotary_pos_emb = rotary_pos_emb.transpose(0, 1).contiguous()
+
+        # Run encoder.
+        hidden_states, presents, all_hidden_states, all_self_attentions = self.encoder(
+            inputs_embeds, attention_mask, rotary_pos_emb=rotary_pos_emb,
+            kv_caches=past_key_values, use_cache=use_cache, output_hidden_states=output_hidden_states
+        )
+
+        if not return_dict:
+            return tuple(v for v in [hidden_states, presents, all_hidden_states, all_self_attentions] if v is not None)
+
+        return BaseModelOutputWithPast(
+            last_hidden_state=hidden_states,
+            past_key_values=presents,
+            hidden_states=all_hidden_states,
+            attentions=all_self_attentions,
+        )
+
+
+class ChatGLMForConditionalGeneration(ChatGLMPreTrainedModel):
+    def __init__(self, config: ChatGLMConfig, empty_init=True, device=None):
+        super().__init__(config)
+
+        self.max_sequence_length = config.max_length
+        self.transformer = ChatGLMModel(config, empty_init=empty_init, device=device)
+        self.config = config
+        self.pack_loss = False
+
+    def _update_model_kwargs_for_generation(
+            self,
+            outputs: ModelOutput,
+            model_kwargs: Dict[str, Any],
+            is_encoder_decoder: bool = False,
+    ) -> Dict[str, Any]:
+        # update past_key_values
+        cache_name, cache = self._extract_past_from_model_output(outputs)
+        model_kwargs[cache_name] = cache
+
+        # update attention mask
+        if "attention_mask" in model_kwargs:
+            attention_mask = model_kwargs["attention_mask"]
+            model_kwargs["attention_mask"] = torch.cat(
+                [attention_mask, attention_mask.new_ones((attention_mask.shape[0], 1))], dim=-1
+            )
+
+        # update position ids
+        if "position_ids" in model_kwargs:
+            position_ids = model_kwargs["position_ids"]
+            new_position_id = position_ids[..., -1:].clone()
+            new_position_id += 1
+            model_kwargs["position_ids"] = torch.cat(
+                [position_ids, new_position_id], dim=-1
+            )
+
+        model_kwargs["is_first_forward"] = False
+        return model_kwargs
+
+    def prepare_inputs_for_generation(
+            self,
+            input_ids: torch.LongTensor,
+            past_key_values: Optional[torch.Tensor] = None,
+            attention_mask: Optional[torch.Tensor] = None,
+            position_ids: Optional[torch.Tensor] = None,
+            is_first_forward: bool = True,
+            **kwargs
+    ) -> dict:
+        # only last token for input_ids if past is not None
+        if position_ids is None:
+            position_ids = self.get_position_ids(input_ids, device=input_ids.device)
+        if not is_first_forward:
+            position_ids = position_ids[..., -1:]
+            input_ids = input_ids[:, -1:]
+        return {
+            "input_ids": input_ids,
+            "past_key_values": past_key_values,
+            "position_ids": position_ids,
+            "attention_mask": attention_mask,
+            "return_last_logit": True
+        }
+
+    def forward(
+            self,
+            input_ids: Optional[torch.Tensor] = None,
+            position_ids: Optional[torch.Tensor] = None,
+            attention_mask: Optional[torch.Tensor] = None,
+            past_key_values: Optional[Tuple[torch.FloatTensor]] = None,
+            inputs_embeds: Optional[torch.Tensor] = None,
+            labels: Optional[Tuple[torch.Tensor]] = None,
+            use_cache: Optional[bool] = None,
+            output_attentions: Optional[bool] = None,
+            output_hidden_states: Optional[bool] = None,
+            return_dict: Optional[bool] = None,
+            return_last_logit: Optional[bool] = False,
+    ):
+        use_cache = use_cache if use_cache is not None else self.config.use_cache
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        transformer_outputs = self.transformer(
+            input_ids=input_ids,
+            position_ids=position_ids,
+            attention_mask=attention_mask,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        hidden_states = transformer_outputs[0]
+        if return_last_logit:
+            hidden_states = hidden_states[-1:]
+        lm_logits = self.transformer.output_layer(hidden_states)
+        lm_logits = lm_logits.transpose(0, 1).contiguous()
+
+        loss = None
+        if labels is not None:
+            lm_logits = lm_logits.to(torch.float32)
+            # Shift so that tokens < n predict n
+            shift_logits = lm_logits[..., :-1, :].contiguous()
+            if isinstance(labels, tuple) or isinstance(labels, list):
+                labels, weights = labels
+            shift_labels = labels[..., 1:].contiguous()
+            if self.pack_loss:
+                loss_fct = CrossEntropyLoss(ignore_index=-100)#, reduction='none')
+                loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1))
+                loss *= weights
+            # if self.pack_loss:
+            #     shift_weights = weights[..., 1:].contiguous()
+            #     loss_fct = CrossEntropyLoss(ignore_index=-100, reduction='none')
+            #     loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1))
+            #     loss = (loss * shift_weights).sum()
+            else:
+                loss_fct = CrossEntropyLoss(ignore_index=-100)
+                loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1))
+
+            lm_logits = lm_logits.to(hidden_states.dtype)
+            loss = loss.to(hidden_states.dtype)
+
+        if not return_dict:
+            output = (lm_logits,) + transformer_outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return CausalLMOutputWithPast(
+            loss=loss,
+            logits=lm_logits,
+            past_key_values=transformer_outputs.past_key_values,
+            hidden_states=transformer_outputs.hidden_states,
+            attentions=transformer_outputs.attentions,
+        )
+
+    @staticmethod
+    def _reorder_cache(
+            past: Tuple[Tuple[torch.Tensor, torch.Tensor], ...], beam_idx: torch.LongTensor
+    ) -> Tuple[Tuple[torch.Tensor, torch.Tensor], ...]:
+        """
+        This function is used to re-order the `past_key_values` cache if [`~PreTrainedModel.beam_search`] or
+        [`~PreTrainedModel.beam_sample`] is called. This is required to match `past_key_values` with the correct
+        beam_idx at every generation step.
+
+        Output shares the same memory storage as `past`.
+        """
+        return tuple(
+            (
+                layer_past[0].index_select(1, beam_idx.to(layer_past[0].device)),
+                layer_past[1].index_select(1, beam_idx.to(layer_past[1].device)),
+            )
+            for layer_past in past
+        )
+
+    def process_response(self, response):
+        response = response.strip()
+        response = response.replace("[[训练时间]]", "2023年")
+        return response
+
+    @torch.inference_mode()
+    def chat(self, tokenizer, query: str, history: List[Dict] = None, role: str = "user",
+             max_length: int = 8192, num_beams=1, do_sample=True, top_p=0.8, temperature=0.8, logits_processor=None,
+             **kwargs):
+        if history is None:
+            history = []
+        if logits_processor is None:
+            logits_processor = LogitsProcessorList()
+        logits_processor.append(InvalidScoreLogitsProcessor())
+        gen_kwargs = {"max_length": max_length, "num_beams": num_beams, "do_sample": do_sample, "top_p": top_p,
+                      "temperature": temperature, "logits_processor": logits_processor, **kwargs}
+        inputs = tokenizer.build_chat_input(query, history=history, role=role)
+        inputs = inputs.to(self.device)
+        eos_token_id = [tokenizer.eos_token_id, tokenizer.get_command("<|user|>"),
+                        tokenizer.get_command("<|observation|>")]
+        outputs = self.generate(**inputs, **gen_kwargs, eos_token_id=eos_token_id)
+        outputs = outputs.tolist()[0][len(inputs["input_ids"][0]):-1]
+        response = tokenizer.decode(outputs)
+        history.append({"role": role, "content": query})
+        response = self.process_response(response)
+        return response, history

smash_config.json

@@ -0,0 +1,31 @@
+{
+    "api_key": null,
+    "verify_url": "http://johnrachwan.pythonanywhere.com",
+    "smash_config": {
+        "pruners": "None",
+        "pruning_ratio": 0.0,
+        "factorizers": "None",
+        "quantizers": "['llm-int8']",
+        "weight_quantization_bits": 4,
+        "output_deviation": 0.005,
+        "compilers": "None",
+        "static_batch": true,
+        "static_shape": true,
+        "controlnet": "None",
+        "unet_dim": 4,
+        "device": "cuda",
+        "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelsdhkyhu_n",
+        "batch_size": 1,
+        "model_name": "THUDM/LongWriter-glm4-9b",
+        "task": "text_text_generation",
+        "max_batch_size": 1,
+        "qtype_weight": "torch.qint8",
+        "qtype_activation": "torch.quint8",
+        "qobserver": "<class 'torch.ao.quantization.observer.MinMaxObserver'>",
+        "qscheme": "torch.per_tensor_symmetric",
+        "qconfig": "x86",
+        "group_size": 128,
+        "damp_percent": 0.1,
+        "save_load_fn": "bitsandbytes"
+    }
+}

special_tokens_map.json

@@ -0,0 +1,9 @@
+{
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a493598071550244b2ee7f26118f3edec2150b9dfa967929a99052ac83fe716
+size 2623634

tokenizer_config.json

@@ -0,0 +1,27 @@
+{
+  "added_tokens_decoder": {
+    "151329": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "auto_map": {
+    "AutoTokenizer": [
+      "THUDM/LongWriter-glm4-9b--tokenization_chatglm.ChatGLM4Tokenizer",
+      null
+    ]
+  },
+  "chat_template": "{% for message in messages %}{% if loop.first %}[gMASK]sop<|{{ message['role'] }}|>\n {{ message['content'] }}{% else %}<|{{ message['role'] }}|>\n {{ message['content'] }}{% endif %}{% endfor %}{% if add_generation_prompt %}<|assistant|>{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "do_lower_case": false,
+  "eos_token": "<|endoftext|>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "padding_side": "left",
+  "remove_space": false,
+  "tokenizer_class": "ChatGLM4Tokenizer"
+}