Upload folder using huggingface_hub

.gitattributes

@@ -43,3 +43,13 @@ google/gemma_2b_scotus/checkpoint-300/tokenizer.json filter=lfs diff=lfs merge=l
 google/gemma_2b_scotus/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 google/gemma_2b_twitter/checkpoint-250/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 google/gemma_2b_twitter/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+LoRA/google/gemma_7b_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-750/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+LoRA/google/gemma_7b_LoRA_MAdAiLab/amazon_attrprompt/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+LoRA/google/gemma_7b_LoRA_MAdAiLab/twitter_disaster/checkpoint-250/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+LoRA/google/gemma_7b_LoRA_MAdAiLab/twitter_disaster/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+LoRA/google/gemma_7b_LoRA_ccdv/patent_classification_abstract/checkpoint-1400/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+LoRA/google/gemma_7b_LoRA_ccdv/patent_classification_abstract/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+LoRA/google/gemma_7b_LoRA_coastalcph/lex_glue_ledgar/checkpoint-3700/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+LoRA/google/gemma_7b_LoRA_coastalcph/lex_glue_ledgar/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+LoRA/google/gemma_7b_LoRA_coastalcph/lex_glue_scotus/checkpoint-450/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+LoRA/google/gemma_7b_LoRA_coastalcph/lex_glue_scotus/tokenizer.json filter=lfs diff=lfs merge=lfs -text

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/README.md

@@ -0,0 +1,89 @@
+---
+license: other
+library_name: peft
+tags:
+- generated_from_trainer
+base_model: Qwen/Qwen1.5-7B
+metrics:
+- accuracy
+model-index:
+- name: amazon_attrprompt
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# amazon_attrprompt
+
+This model is a fine-tuned version of [Qwen/Qwen1.5-7B](https://huggingface.co/Qwen/Qwen1.5-7B) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4250
+- Accuracy: 0.8709
+- F1 Macro: 0.8541
+- F1 Micro: 0.8709
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- total_train_batch_size: 32
+- total_eval_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 3.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:|
+| 1.6833        | 0.13  | 50   | 1.2279          | 0.6640   | 0.5879   | 0.6640   |
+| 0.6531        | 0.26  | 100  | 0.6578          | 0.8155   | 0.7767   | 0.8155   |
+| 0.6075        | 0.39  | 150  | 0.5935          | 0.8327   | 0.8113   | 0.8327   |
+| 0.5646        | 0.53  | 200  | 0.5660          | 0.8379   | 0.8194   | 0.8379   |
+| 0.6148        | 0.66  | 250  | 0.5318          | 0.8426   | 0.8319   | 0.8426   |
+| 0.4047        | 0.79  | 300  | 0.4546          | 0.8650   | 0.8467   | 0.8650   |
+| 0.568         | 0.92  | 350  | 0.4250          | 0.8709   | 0.8541   | 0.8709   |
+| 0.2395        | 1.05  | 400  | 0.4570          | 0.8762   | 0.8611   | 0.8762   |
+| 0.2213        | 1.18  | 450  | 0.4524          | 0.8775   | 0.8631   | 0.8775   |
+| 0.1778        | 1.32  | 500  | 0.4649          | 0.8748   | 0.8508   | 0.8748   |
+| 0.1738        | 1.45  | 550  | 0.4853          | 0.8794   | 0.8617   | 0.8794   |
+| 0.2643        | 1.58  | 600  | 0.4302          | 0.8827   | 0.8676   | 0.8827   |
+| 0.3357        | 1.71  | 650  | 0.4388          | 0.8827   | 0.8673   | 0.8827   |
+| 0.3029        | 1.84  | 700  | 0.4431          | 0.8827   | 0.8656   | 0.8827   |
+| 0.1809        | 1.97  | 750  | 0.4266          | 0.8900   | 0.8742   | 0.8900   |
+| 0.0589        | 2.11  | 800  | 0.4499          | 0.8946   | 0.8815   | 0.8946   |
+| 0.0531        | 2.24  | 850  | 0.4758          | 0.8920   | 0.8758   | 0.8920   |
+| 0.0234        | 2.37  | 900  | 0.4788          | 0.8953   | 0.8804   | 0.8953   |
+| 0.0145        | 2.5   | 950  | 0.4976          | 0.8939   | 0.8779   | 0.8939   |
+| 0.058         | 2.63  | 1000 | 0.4967          | 0.8992   | 0.8816   | 0.8992   |
+| 0.05          | 2.76  | 1050 | 0.5113          | 0.8933   | 0.8753   | 0.8933   |
+| 0.0556        | 2.89  | 1100 | 0.5024          | 0.8966   | 0.8803   | 0.8966   |
+
+
+### Framework versions
+
+- PEFT 0.9.0
+- Transformers 4.39.0.dev0
+- Pytorch 2.2.1+cu121
+- Datasets 2.18.0
+- Tokenizers 0.15.2

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/adapter_config.json

@@ -0,0 +1,33 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen1.5-7B",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 256,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "q_proj",
+    "down_proj",
+    "k_proj",
+    "up_proj",
+    "gate_proj"
+  ],
+  "task_type": "SEQ_CLS",
+  "use_dora": false,
+  "use_rslora": false
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd7a062e5f54abbe183ffc004af744d6b392306c42e0a9361e709f5ee210f6a1
+size 1882260178

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/all_results.json

@@ -0,0 +1,23 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.8708827404479579,
+    "eval_f1_macro": 0.8540867659285116,
+    "eval_f1_micro": 0.8708827404479579,
+    "eval_loss": 0.42495009303092957,
+    "eval_runtime": 18.5363,
+    "eval_samples": 1518,
+    "eval_samples_per_second": 81.893,
+    "eval_steps_per_second": 2.59,
+    "test_accuracy": 0.8787878787878788,
+    "test_f1_macro": 0.8592526000275965,
+    "test_f1_micro": 0.8787878787878788,
+    "test_loss": 0.4176747500896454,
+    "test_runtime": 18.7826,
+    "test_samples_per_second": 80.82,
+    "test_steps_per_second": 2.556,
+    "train_loss": 0.48372833394167714,
+    "train_runtime": 1860.5106,
+    "train_samples": 12144,
+    "train_samples_per_second": 19.582,
+    "train_steps_per_second": 0.613
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/README.md

@@ -0,0 +1,202 @@
+---
+library_name: peft
+base_model: Qwen/Qwen1.5-7B
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.9.0

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/adapter_config.json

@@ -0,0 +1,33 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen1.5-7B",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 256,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "q_proj",
+    "down_proj",
+    "k_proj",
+    "up_proj",
+    "gate_proj"
+  ],
+  "task_type": "SEQ_CLS",
+  "use_dora": false,
+  "use_rslora": false
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd7a062e5f54abbe183ffc004af744d6b392306c42e0a9361e709f5ee210f6a1
+size 1882260178

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/global_step350/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1331ee472a3bb3b5c75befdbaa87ce114569684e7e2dc4aaba5be969d65aa55
+size 1919487472

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/global_step350/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:556daff7ae35ddc09de810d220383d9c830438789751d565e2afa1289f8eb12d
+size 1919487600

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/global_step350/mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:533e7a2b9d300977594023b037f52411c969287ffccd2ca72745454822bcedfe
+size 640146476

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/latest

@@ -0,0 +1 @@
+global_step350

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1bf511f6e62a4ac4d0dcff2abdca4aafd83da0347b48962eb7a4450af95587f
+size 14512

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6f7e6176606e470603dae75d8a0b1345f81b06321b18f2dac42a33fe19eec3b
+size 14512

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72c981663971b3602dc2dedc7aef262e19f3e64556be6dd4800b987af7759a5b
+size 1064

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/special_tokens_map.json

@@ -0,0 +1,14 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>"
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/tokenizer_config.json

@@ -0,0 +1,43 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/trainer_state.json

@@ -0,0 +1,343 @@
+{
+  "best_metric": 0.42495009303092957,
+  "best_model_checkpoint": "../experiments_checkpoints/LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350",
+  "epoch": 0.9210526315789473,
+  "eval_steps": 50,
+  "global_step": 350,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "grad_norm": 88.5622329711914,
+      "learning_rate": 4.956140350877193e-05,
+      "loss": 9.8375,
+      "step": 10
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 76.16983032226562,
+      "learning_rate": 4.912280701754386e-05,
+      "loss": 6.3422,
+      "step": 20
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 68.94097137451172,
+      "learning_rate": 4.868421052631579e-05,
+      "loss": 3.866,
+      "step": 30
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 72.53247833251953,
+      "learning_rate": 4.824561403508772e-05,
+      "loss": 2.1033,
+      "step": 40
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 67.55592346191406,
+      "learning_rate": 4.780701754385965e-05,
+      "loss": 1.6833,
+      "step": 50
+    },
+    {
+      "epoch": 0.13,
+      "eval_accuracy": 0.6640316205533597,
+      "eval_f1_macro": 0.5878963576218975,
+      "eval_f1_micro": 0.6640316205533597,
+      "eval_loss": 1.2279417514801025,
+      "eval_runtime": 19.0551,
+      "eval_samples_per_second": 79.664,
+      "eval_steps_per_second": 2.519,
+      "step": 50
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 34.1900520324707,
+      "learning_rate": 4.736842105263158e-05,
+      "loss": 1.0584,
+      "step": 60
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 48.36394119262695,
+      "learning_rate": 4.6929824561403515e-05,
+      "loss": 1.0168,
+      "step": 70
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 46.12382888793945,
+      "learning_rate": 4.649122807017544e-05,
+      "loss": 0.8845,
+      "step": 80
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 43.66932678222656,
+      "learning_rate": 4.605263157894737e-05,
+      "loss": 0.8027,
+      "step": 90
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 28.10245132446289,
+      "learning_rate": 4.56140350877193e-05,
+      "loss": 0.6531,
+      "step": 100
+    },
+    {
+      "epoch": 0.26,
+      "eval_accuracy": 0.8155467720685112,
+      "eval_f1_macro": 0.7766571443103144,
+      "eval_f1_micro": 0.8155467720685112,
+      "eval_loss": 0.6577733755111694,
+      "eval_runtime": 19.1186,
+      "eval_samples_per_second": 79.399,
+      "eval_steps_per_second": 2.511,
+      "step": 100
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 59.69050216674805,
+      "learning_rate": 4.517543859649123e-05,
+      "loss": 0.8677,
+      "step": 110
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 56.882625579833984,
+      "learning_rate": 4.473684210526316e-05,
+      "loss": 0.7174,
+      "step": 120
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 29.96848487854004,
+      "learning_rate": 4.429824561403509e-05,
+      "loss": 0.5984,
+      "step": 130
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 53.101444244384766,
+      "learning_rate": 4.3859649122807014e-05,
+      "loss": 0.5318,
+      "step": 140
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 30.550228118896484,
+      "learning_rate": 4.342105263157895e-05,
+      "loss": 0.6075,
+      "step": 150
+    },
+    {
+      "epoch": 0.39,
+      "eval_accuracy": 0.8326745718050066,
+      "eval_f1_macro": 0.8113232244484544,
+      "eval_f1_micro": 0.8326745718050066,
+      "eval_loss": 0.5934926867485046,
+      "eval_runtime": 19.1445,
+      "eval_samples_per_second": 79.292,
+      "eval_steps_per_second": 2.507,
+      "step": 150
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 47.42914581298828,
+      "learning_rate": 4.298245614035088e-05,
+      "loss": 0.7186,
+      "step": 160
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 34.166526794433594,
+      "learning_rate": 4.254385964912281e-05,
+      "loss": 0.7026,
+      "step": 170
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 28.214841842651367,
+      "learning_rate": 4.210526315789474e-05,
+      "loss": 0.6292,
+      "step": 180
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 45.511444091796875,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.7115,
+      "step": 190
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 18.114608764648438,
+      "learning_rate": 4.12280701754386e-05,
+      "loss": 0.5646,
+      "step": 200
+    },
+    {
+      "epoch": 0.53,
+      "eval_accuracy": 0.8379446640316206,
+      "eval_f1_macro": 0.8193854450487876,
+      "eval_f1_micro": 0.8379446640316206,
+      "eval_loss": 0.5659688115119934,
+      "eval_runtime": 19.1468,
+      "eval_samples_per_second": 79.282,
+      "eval_steps_per_second": 2.507,
+      "step": 200
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 29.09646987915039,
+      "learning_rate": 4.078947368421053e-05,
+      "loss": 0.4866,
+      "step": 210
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 44.29309844970703,
+      "learning_rate": 4.0350877192982455e-05,
+      "loss": 0.4314,
+      "step": 220
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 42.93623733520508,
+      "learning_rate": 3.991228070175439e-05,
+      "loss": 0.7159,
+      "step": 230
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 33.71025848388672,
+      "learning_rate": 3.9473684210526316e-05,
+      "loss": 0.6136,
+      "step": 240
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 42.2908821105957,
+      "learning_rate": 3.9035087719298244e-05,
+      "loss": 0.6148,
+      "step": 250
+    },
+    {
+      "epoch": 0.66,
+      "eval_accuracy": 0.8425559947299077,
+      "eval_f1_macro": 0.8319363334583545,
+      "eval_f1_micro": 0.8425559947299077,
+      "eval_loss": 0.531785249710083,
+      "eval_runtime": 19.145,
+      "eval_samples_per_second": 79.29,
+      "eval_steps_per_second": 2.507,
+      "step": 250
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 38.71971130371094,
+      "learning_rate": 3.859649122807018e-05,
+      "loss": 0.582,
+      "step": 260
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 21.194923400878906,
+      "learning_rate": 3.815789473684211e-05,
+      "loss": 0.5039,
+      "step": 270
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 47.179935455322266,
+      "learning_rate": 3.771929824561404e-05,
+      "loss": 0.5564,
+      "step": 280
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 34.193275451660156,
+      "learning_rate": 3.728070175438597e-05,
+      "loss": 0.452,
+      "step": 290
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 20.546531677246094,
+      "learning_rate": 3.6842105263157895e-05,
+      "loss": 0.4047,
+      "step": 300
+    },
+    {
+      "epoch": 0.79,
+      "eval_accuracy": 0.8649538866930171,
+      "eval_f1_macro": 0.8467453346863307,
+      "eval_f1_micro": 0.8649538866930171,
+      "eval_loss": 0.4545969069004059,
+      "eval_runtime": 19.151,
+      "eval_samples_per_second": 79.265,
+      "eval_steps_per_second": 2.506,
+      "step": 300
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 46.09516525268555,
+      "learning_rate": 3.640350877192983e-05,
+      "loss": 0.4348,
+      "step": 310
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 35.31401443481445,
+      "learning_rate": 3.5964912280701756e-05,
+      "loss": 0.5397,
+      "step": 320
+    },
+    {
+      "epoch": 0.87,
+      "grad_norm": 22.361942291259766,
+      "learning_rate": 3.5526315789473684e-05,
+      "loss": 0.5323,
+      "step": 330
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 42.461910247802734,
+      "learning_rate": 3.508771929824561e-05,
+      "loss": 0.5028,
+      "step": 340
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 35.93864059448242,
+      "learning_rate": 3.4649122807017546e-05,
+      "loss": 0.568,
+      "step": 350
+    },
+    {
+      "epoch": 0.92,
+      "eval_accuracy": 0.8708827404479579,
+      "eval_f1_macro": 0.8540867659285116,
+      "eval_f1_micro": 0.8708827404479579,
+      "eval_loss": 0.42495009303092957,
+      "eval_runtime": 19.1783,
+      "eval_samples_per_second": 79.152,
+      "eval_steps_per_second": 2.503,
+      "step": 350
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 1140,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 50,
+  "total_flos": 5.846222510227456e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95b4a5adabb844b5e8347d17e6ffdbb68778d8d70412e41fe9f46cfcf73b127d
+size 6008

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350/zero_to_fp32.py

@@ -0,0 +1,604 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example: python zero_to_fp32.py . pytorch_model.bin
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in param_shapes.items():
+
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
+    print(f"Saving fp32 state dict to {output_file}")
+    torch.save(state_dict, output_file)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument(
+        "output_file",
+        type=str,
+        help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_file,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.8708827404479579,
+    "eval_f1_macro": 0.8540867659285116,
+    "eval_f1_micro": 0.8708827404479579,
+    "eval_loss": 0.42495009303092957,
+    "eval_runtime": 18.5363,
+    "eval_samples": 1518,
+    "eval_samples_per_second": 81.893,
+    "eval_steps_per_second": 2.59
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/run.log

@@ -0,0 +1,4 @@
+03/18/2024 19:06:24 - WARNING - __main__ - Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: True, 16-bits training: False
+03/18/2024 19:06:24 - WARNING - __main__ - Process rank: 1, device: cuda:1, n_gpu: 1, distributed training: True, 16-bits training: False
+03/18/2024 19:06:46 - WARNING - __main__ - The label2id key in the model config.json is not equal to the label2id key of this run. You can ignore this if you are doing finetuning.
+03/18/2024 19:06:46 - WARNING - __main__ - The label2id key in the model config.json is not equal to the label2id key of this run. You can ignore this if you are doing finetuning.

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/special_tokens_map.json

@@ -0,0 +1,14 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>"
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/test_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 3.0,
+    "test_accuracy": 0.8787878787878788,
+    "test_f1_macro": 0.8592526000275965,
+    "test_f1_micro": 0.8787878787878788,
+    "test_loss": 0.4176747500896454,
+    "test_runtime": 18.7826,
+    "test_samples_per_second": 80.82,
+    "test_steps_per_second": 2.556
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/tokenizer_config.json

@@ -0,0 +1,43 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 0.48372833394167714,
+    "train_runtime": 1860.5106,
+    "train_samples": 12144,
+    "train_samples_per_second": 19.582,
+    "train_steps_per_second": 0.613
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/trainer_state.json

@@ -0,0 +1,1070 @@
+{
+  "best_metric": 0.42495009303092957,
+  "best_model_checkpoint": "../experiments_checkpoints/LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/checkpoint-350",
+  "epoch": 3.0,
+  "eval_steps": 50,
+  "global_step": 1140,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "grad_norm": 88.5622329711914,
+      "learning_rate": 4.956140350877193e-05,
+      "loss": 9.8375,
+      "step": 10
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 76.16983032226562,
+      "learning_rate": 4.912280701754386e-05,
+      "loss": 6.3422,
+      "step": 20
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 68.94097137451172,
+      "learning_rate": 4.868421052631579e-05,
+      "loss": 3.866,
+      "step": 30
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 72.53247833251953,
+      "learning_rate": 4.824561403508772e-05,
+      "loss": 2.1033,
+      "step": 40
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 67.55592346191406,
+      "learning_rate": 4.780701754385965e-05,
+      "loss": 1.6833,
+      "step": 50
+    },
+    {
+      "epoch": 0.13,
+      "eval_accuracy": 0.6640316205533597,
+      "eval_f1_macro": 0.5878963576218975,
+      "eval_f1_micro": 0.6640316205533597,
+      "eval_loss": 1.2279417514801025,
+      "eval_runtime": 19.0551,
+      "eval_samples_per_second": 79.664,
+      "eval_steps_per_second": 2.519,
+      "step": 50
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 34.1900520324707,
+      "learning_rate": 4.736842105263158e-05,
+      "loss": 1.0584,
+      "step": 60
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 48.36394119262695,
+      "learning_rate": 4.6929824561403515e-05,
+      "loss": 1.0168,
+      "step": 70
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 46.12382888793945,
+      "learning_rate": 4.649122807017544e-05,
+      "loss": 0.8845,
+      "step": 80
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 43.66932678222656,
+      "learning_rate": 4.605263157894737e-05,
+      "loss": 0.8027,
+      "step": 90
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 28.10245132446289,
+      "learning_rate": 4.56140350877193e-05,
+      "loss": 0.6531,
+      "step": 100
+    },
+    {
+      "epoch": 0.26,
+      "eval_accuracy": 0.8155467720685112,
+      "eval_f1_macro": 0.7766571443103144,
+      "eval_f1_micro": 0.8155467720685112,
+      "eval_loss": 0.6577733755111694,
+      "eval_runtime": 19.1186,
+      "eval_samples_per_second": 79.399,
+      "eval_steps_per_second": 2.511,
+      "step": 100
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 59.69050216674805,
+      "learning_rate": 4.517543859649123e-05,
+      "loss": 0.8677,
+      "step": 110
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 56.882625579833984,
+      "learning_rate": 4.473684210526316e-05,
+      "loss": 0.7174,
+      "step": 120
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 29.96848487854004,
+      "learning_rate": 4.429824561403509e-05,
+      "loss": 0.5984,
+      "step": 130
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 53.101444244384766,
+      "learning_rate": 4.3859649122807014e-05,
+      "loss": 0.5318,
+      "step": 140
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 30.550228118896484,
+      "learning_rate": 4.342105263157895e-05,
+      "loss": 0.6075,
+      "step": 150
+    },
+    {
+      "epoch": 0.39,
+      "eval_accuracy": 0.8326745718050066,
+      "eval_f1_macro": 0.8113232244484544,
+      "eval_f1_micro": 0.8326745718050066,
+      "eval_loss": 0.5934926867485046,
+      "eval_runtime": 19.1445,
+      "eval_samples_per_second": 79.292,
+      "eval_steps_per_second": 2.507,
+      "step": 150
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 47.42914581298828,
+      "learning_rate": 4.298245614035088e-05,
+      "loss": 0.7186,
+      "step": 160
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 34.166526794433594,
+      "learning_rate": 4.254385964912281e-05,
+      "loss": 0.7026,
+      "step": 170
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 28.214841842651367,
+      "learning_rate": 4.210526315789474e-05,
+      "loss": 0.6292,
+      "step": 180
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 45.511444091796875,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.7115,
+      "step": 190
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 18.114608764648438,
+      "learning_rate": 4.12280701754386e-05,
+      "loss": 0.5646,
+      "step": 200
+    },
+    {
+      "epoch": 0.53,
+      "eval_accuracy": 0.8379446640316206,
+      "eval_f1_macro": 0.8193854450487876,
+      "eval_f1_micro": 0.8379446640316206,
+      "eval_loss": 0.5659688115119934,
+      "eval_runtime": 19.1468,
+      "eval_samples_per_second": 79.282,
+      "eval_steps_per_second": 2.507,
+      "step": 200
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 29.09646987915039,
+      "learning_rate": 4.078947368421053e-05,
+      "loss": 0.4866,
+      "step": 210
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 44.29309844970703,
+      "learning_rate": 4.0350877192982455e-05,
+      "loss": 0.4314,
+      "step": 220
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 42.93623733520508,
+      "learning_rate": 3.991228070175439e-05,
+      "loss": 0.7159,
+      "step": 230
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 33.71025848388672,
+      "learning_rate": 3.9473684210526316e-05,
+      "loss": 0.6136,
+      "step": 240
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 42.2908821105957,
+      "learning_rate": 3.9035087719298244e-05,
+      "loss": 0.6148,
+      "step": 250
+    },
+    {
+      "epoch": 0.66,
+      "eval_accuracy": 0.8425559947299077,
+      "eval_f1_macro": 0.8319363334583545,
+      "eval_f1_micro": 0.8425559947299077,
+      "eval_loss": 0.531785249710083,
+      "eval_runtime": 19.145,
+      "eval_samples_per_second": 79.29,
+      "eval_steps_per_second": 2.507,
+      "step": 250
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 38.71971130371094,
+      "learning_rate": 3.859649122807018e-05,
+      "loss": 0.582,
+      "step": 260
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 21.194923400878906,
+      "learning_rate": 3.815789473684211e-05,
+      "loss": 0.5039,
+      "step": 270
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 47.179935455322266,
+      "learning_rate": 3.771929824561404e-05,
+      "loss": 0.5564,
+      "step": 280
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 34.193275451660156,
+      "learning_rate": 3.728070175438597e-05,
+      "loss": 0.452,
+      "step": 290
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 20.546531677246094,
+      "learning_rate": 3.6842105263157895e-05,
+      "loss": 0.4047,
+      "step": 300
+    },
+    {
+      "epoch": 0.79,
+      "eval_accuracy": 0.8649538866930171,
+      "eval_f1_macro": 0.8467453346863307,
+      "eval_f1_micro": 0.8649538866930171,
+      "eval_loss": 0.4545969069004059,
+      "eval_runtime": 19.151,
+      "eval_samples_per_second": 79.265,
+      "eval_steps_per_second": 2.506,
+      "step": 300
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 46.09516525268555,
+      "learning_rate": 3.640350877192983e-05,
+      "loss": 0.4348,
+      "step": 310
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 35.31401443481445,
+      "learning_rate": 3.5964912280701756e-05,
+      "loss": 0.5397,
+      "step": 320
+    },
+    {
+      "epoch": 0.87,
+      "grad_norm": 22.361942291259766,
+      "learning_rate": 3.5526315789473684e-05,
+      "loss": 0.5323,
+      "step": 330
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 42.461910247802734,
+      "learning_rate": 3.508771929824561e-05,
+      "loss": 0.5028,
+      "step": 340
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 35.93864059448242,
+      "learning_rate": 3.4649122807017546e-05,
+      "loss": 0.568,
+      "step": 350
+    },
+    {
+      "epoch": 0.92,
+      "eval_accuracy": 0.8708827404479579,
+      "eval_f1_macro": 0.8540867659285116,
+      "eval_f1_micro": 0.8708827404479579,
+      "eval_loss": 0.42495009303092957,
+      "eval_runtime": 19.1783,
+      "eval_samples_per_second": 79.152,
+      "eval_steps_per_second": 2.503,
+      "step": 350
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 19.8919734954834,
+      "learning_rate": 3.421052631578947e-05,
+      "loss": 0.3863,
+      "step": 360
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 46.49361801147461,
+      "learning_rate": 3.377192982456141e-05,
+      "loss": 0.5599,
+      "step": 370
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 31.97792625427246,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.5042,
+      "step": 380
+    },
+    {
+      "epoch": 1.03,
+      "grad_norm": 30.975046157836914,
+      "learning_rate": 3.289473684210527e-05,
+      "loss": 0.3024,
+      "step": 390
+    },
+    {
+      "epoch": 1.05,
+      "grad_norm": 14.051041603088379,
+      "learning_rate": 3.24561403508772e-05,
+      "loss": 0.2395,
+      "step": 400
+    },
+    {
+      "epoch": 1.05,
+      "eval_accuracy": 0.8761528326745718,
+      "eval_f1_macro": 0.8611335072766768,
+      "eval_f1_micro": 0.8761528326745718,
+      "eval_loss": 0.4569604992866516,
+      "eval_runtime": 19.1561,
+      "eval_samples_per_second": 79.244,
+      "eval_steps_per_second": 2.506,
+      "step": 400
+    },
+    {
+      "epoch": 1.08,
+      "grad_norm": 28.197834014892578,
+      "learning_rate": 3.2017543859649124e-05,
+      "loss": 0.2336,
+      "step": 410
+    },
+    {
+      "epoch": 1.11,
+      "grad_norm": 17.73228645324707,
+      "learning_rate": 3.157894736842105e-05,
+      "loss": 0.2069,
+      "step": 420
+    },
+    {
+      "epoch": 1.13,
+      "grad_norm": 19.042102813720703,
+      "learning_rate": 3.1140350877192986e-05,
+      "loss": 0.3182,
+      "step": 430
+    },
+    {
+      "epoch": 1.16,
+      "grad_norm": 17.37108039855957,
+      "learning_rate": 3.0701754385964913e-05,
+      "loss": 0.2071,
+      "step": 440
+    },
+    {
+      "epoch": 1.18,
+      "grad_norm": 17.802289962768555,
+      "learning_rate": 3.0263157894736844e-05,
+      "loss": 0.2213,
+      "step": 450
+    },
+    {
+      "epoch": 1.18,
+      "eval_accuracy": 0.8774703557312253,
+      "eval_f1_macro": 0.8631213514149485,
+      "eval_f1_micro": 0.8774703557312253,
+      "eval_loss": 0.45242956280708313,
+      "eval_runtime": 19.1508,
+      "eval_samples_per_second": 79.266,
+      "eval_steps_per_second": 2.506,
+      "step": 450
+    },
+    {
+      "epoch": 1.21,
+      "grad_norm": 18.258426666259766,
+      "learning_rate": 2.9824561403508772e-05,
+      "loss": 0.2722,
+      "step": 460
+    },
+    {
+      "epoch": 1.24,
+      "grad_norm": 24.840511322021484,
+      "learning_rate": 2.9385964912280706e-05,
+      "loss": 0.2721,
+      "step": 470
+    },
+    {
+      "epoch": 1.26,
+      "grad_norm": 14.744382858276367,
+      "learning_rate": 2.8947368421052634e-05,
+      "loss": 0.2501,
+      "step": 480
+    },
+    {
+      "epoch": 1.29,
+      "grad_norm": 18.21392250061035,
+      "learning_rate": 2.850877192982456e-05,
+      "loss": 0.2516,
+      "step": 490
+    },
+    {
+      "epoch": 1.32,
+      "grad_norm": 7.454222679138184,
+      "learning_rate": 2.8070175438596492e-05,
+      "loss": 0.1778,
+      "step": 500
+    },
+    {
+      "epoch": 1.32,
+      "eval_accuracy": 0.8748353096179183,
+      "eval_f1_macro": 0.8507965490074818,
+      "eval_f1_micro": 0.8748353096179183,
+      "eval_loss": 0.4648575782775879,
+      "eval_runtime": 19.1651,
+      "eval_samples_per_second": 79.207,
+      "eval_steps_per_second": 2.505,
+      "step": 500
+    },
+    {
+      "epoch": 1.34,
+      "grad_norm": 15.366374969482422,
+      "learning_rate": 2.7631578947368426e-05,
+      "loss": 0.1821,
+      "step": 510
+    },
+    {
+      "epoch": 1.37,
+      "grad_norm": 17.506078720092773,
+      "learning_rate": 2.7192982456140354e-05,
+      "loss": 0.2568,
+      "step": 520
+    },
+    {
+      "epoch": 1.39,
+      "grad_norm": 32.42897033691406,
+      "learning_rate": 2.675438596491228e-05,
+      "loss": 0.2519,
+      "step": 530
+    },
+    {
+      "epoch": 1.42,
+      "grad_norm": 16.54423713684082,
+      "learning_rate": 2.6315789473684212e-05,
+      "loss": 0.1991,
+      "step": 540
+    },
+    {
+      "epoch": 1.45,
+      "grad_norm": 13.094934463500977,
+      "learning_rate": 2.5877192982456143e-05,
+      "loss": 0.1738,
+      "step": 550
+    },
+    {
+      "epoch": 1.45,
+      "eval_accuracy": 0.8794466403162056,
+      "eval_f1_macro": 0.8616811925870786,
+      "eval_f1_micro": 0.8794466403162056,
+      "eval_loss": 0.4853415787220001,
+      "eval_runtime": 19.1386,
+      "eval_samples_per_second": 79.316,
+      "eval_steps_per_second": 2.508,
+      "step": 550
+    },
+    {
+      "epoch": 1.47,
+      "grad_norm": 9.557943344116211,
+      "learning_rate": 2.5438596491228074e-05,
+      "loss": 0.3554,
+      "step": 560
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 20.867530822753906,
+      "learning_rate": 2.5e-05,
+      "loss": 0.2558,
+      "step": 570
+    },
+    {
+      "epoch": 1.53,
+      "grad_norm": 16.902305603027344,
+      "learning_rate": 2.456140350877193e-05,
+      "loss": 0.2638,
+      "step": 580
+    },
+    {
+      "epoch": 1.55,
+      "grad_norm": 14.376564979553223,
+      "learning_rate": 2.412280701754386e-05,
+      "loss": 0.1744,
+      "step": 590
+    },
+    {
+      "epoch": 1.58,
+      "grad_norm": 21.982099533081055,
+      "learning_rate": 2.368421052631579e-05,
+      "loss": 0.2643,
+      "step": 600
+    },
+    {
+      "epoch": 1.58,
+      "eval_accuracy": 0.8827404479578392,
+      "eval_f1_macro": 0.8675629871629298,
+      "eval_f1_micro": 0.8827404479578392,
+      "eval_loss": 0.43024396896362305,
+      "eval_runtime": 19.1559,
+      "eval_samples_per_second": 79.244,
+      "eval_steps_per_second": 2.506,
+      "step": 600
+    },
+    {
+      "epoch": 1.61,
+      "grad_norm": 8.151387214660645,
+      "learning_rate": 2.324561403508772e-05,
+      "loss": 0.2679,
+      "step": 610
+    },
+    {
+      "epoch": 1.63,
+      "grad_norm": 22.84608268737793,
+      "learning_rate": 2.280701754385965e-05,
+      "loss": 0.2687,
+      "step": 620
+    },
+    {
+      "epoch": 1.66,
+      "grad_norm": 18.665374755859375,
+      "learning_rate": 2.236842105263158e-05,
+      "loss": 0.1821,
+      "step": 630
+    },
+    {
+      "epoch": 1.68,
+      "grad_norm": 16.556060791015625,
+      "learning_rate": 2.1929824561403507e-05,
+      "loss": 0.2583,
+      "step": 640
+    },
+    {
+      "epoch": 1.71,
+      "grad_norm": 10.120057106018066,
+      "learning_rate": 2.149122807017544e-05,
+      "loss": 0.3357,
+      "step": 650
+    },
+    {
+      "epoch": 1.71,
+      "eval_accuracy": 0.8827404479578392,
+      "eval_f1_macro": 0.8673002865339278,
+      "eval_f1_micro": 0.8827404479578392,
+      "eval_loss": 0.43883296847343445,
+      "eval_runtime": 19.1504,
+      "eval_samples_per_second": 79.267,
+      "eval_steps_per_second": 2.506,
+      "step": 650
+    },
+    {
+      "epoch": 1.74,
+      "grad_norm": 27.886629104614258,
+      "learning_rate": 2.105263157894737e-05,
+      "loss": 0.2225,
+      "step": 660
+    },
+    {
+      "epoch": 1.76,
+      "grad_norm": 17.229507446289062,
+      "learning_rate": 2.06140350877193e-05,
+      "loss": 0.2053,
+      "step": 670
+    },
+    {
+      "epoch": 1.79,
+      "grad_norm": 33.89767837524414,
+      "learning_rate": 2.0175438596491227e-05,
+      "loss": 0.223,
+      "step": 680
+    },
+    {
+      "epoch": 1.82,
+      "grad_norm": 12.045727729797363,
+      "learning_rate": 1.9736842105263158e-05,
+      "loss": 0.1707,
+      "step": 690
+    },
+    {
+      "epoch": 1.84,
+      "grad_norm": 19.970233917236328,
+      "learning_rate": 1.929824561403509e-05,
+      "loss": 0.3029,
+      "step": 700
+    },
+    {
+      "epoch": 1.84,
+      "eval_accuracy": 0.8827404479578392,
+      "eval_f1_macro": 0.8655521627258196,
+      "eval_f1_micro": 0.8827404479578392,
+      "eval_loss": 0.4430885314941406,
+      "eval_runtime": 19.1338,
+      "eval_samples_per_second": 79.336,
+      "eval_steps_per_second": 2.509,
+      "step": 700
+    },
+    {
+      "epoch": 1.87,
+      "grad_norm": 16.294740676879883,
+      "learning_rate": 1.885964912280702e-05,
+      "loss": 0.222,
+      "step": 710
+    },
+    {
+      "epoch": 1.89,
+      "grad_norm": 12.050216674804688,
+      "learning_rate": 1.8421052631578947e-05,
+      "loss": 0.2102,
+      "step": 720
+    },
+    {
+      "epoch": 1.92,
+      "grad_norm": 21.319595336914062,
+      "learning_rate": 1.7982456140350878e-05,
+      "loss": 0.1974,
+      "step": 730
+    },
+    {
+      "epoch": 1.95,
+      "grad_norm": 9.788517951965332,
+      "learning_rate": 1.7543859649122806e-05,
+      "loss": 0.2275,
+      "step": 740
+    },
+    {
+      "epoch": 1.97,
+      "grad_norm": 19.825088500976562,
+      "learning_rate": 1.7105263157894737e-05,
+      "loss": 0.1809,
+      "step": 750
+    },
+    {
+      "epoch": 1.97,
+      "eval_accuracy": 0.8899868247694335,
+      "eval_f1_macro": 0.874229892393047,
+      "eval_f1_micro": 0.8899868247694335,
+      "eval_loss": 0.42660534381866455,
+      "eval_runtime": 19.1466,
+      "eval_samples_per_second": 79.283,
+      "eval_steps_per_second": 2.507,
+      "step": 750
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 14.426370620727539,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.1963,
+      "step": 760
+    },
+    {
+      "epoch": 2.03,
+      "grad_norm": 17.27349281311035,
+      "learning_rate": 1.62280701754386e-05,
+      "loss": 0.1066,
+      "step": 770
+    },
+    {
+      "epoch": 2.05,
+      "grad_norm": 3.37688946723938,
+      "learning_rate": 1.5789473684210526e-05,
+      "loss": 0.064,
+      "step": 780
+    },
+    {
+      "epoch": 2.08,
+      "grad_norm": 6.233172416687012,
+      "learning_rate": 1.5350877192982457e-05,
+      "loss": 0.0603,
+      "step": 790
+    },
+    {
+      "epoch": 2.11,
+      "grad_norm": 17.858177185058594,
+      "learning_rate": 1.4912280701754386e-05,
+      "loss": 0.0589,
+      "step": 800
+    },
+    {
+      "epoch": 2.11,
+      "eval_accuracy": 0.8945981554677207,
+      "eval_f1_macro": 0.8814951057007789,
+      "eval_f1_micro": 0.8945981554677207,
+      "eval_loss": 0.4498850703239441,
+      "eval_runtime": 19.1446,
+      "eval_samples_per_second": 79.291,
+      "eval_steps_per_second": 2.507,
+      "step": 800
+    },
+    {
+      "epoch": 2.13,
+      "grad_norm": 10.511329650878906,
+      "learning_rate": 1.4473684210526317e-05,
+      "loss": 0.0526,
+      "step": 810
+    },
+    {
+      "epoch": 2.16,
+      "grad_norm": 3.655291795730591,
+      "learning_rate": 1.4035087719298246e-05,
+      "loss": 0.0308,
+      "step": 820
+    },
+    {
+      "epoch": 2.18,
+      "grad_norm": 14.438605308532715,
+      "learning_rate": 1.3596491228070177e-05,
+      "loss": 0.063,
+      "step": 830
+    },
+    {
+      "epoch": 2.21,
+      "grad_norm": 5.1180195808410645,
+      "learning_rate": 1.3157894736842106e-05,
+      "loss": 0.0297,
+      "step": 840
+    },
+    {
+      "epoch": 2.24,
+      "grad_norm": 9.131390571594238,
+      "learning_rate": 1.2719298245614037e-05,
+      "loss": 0.0531,
+      "step": 850
+    },
+    {
+      "epoch": 2.24,
+      "eval_accuracy": 0.8919631093544137,
+      "eval_f1_macro": 0.8757800595234113,
+      "eval_f1_micro": 0.8919631093544137,
+      "eval_loss": 0.475754976272583,
+      "eval_runtime": 19.1667,
+      "eval_samples_per_second": 79.2,
+      "eval_steps_per_second": 2.504,
+      "step": 850
+    },
+    {
+      "epoch": 2.26,
+      "grad_norm": 0.74644935131073,
+      "learning_rate": 1.2280701754385964e-05,
+      "loss": 0.0649,
+      "step": 860
+    },
+    {
+      "epoch": 2.29,
+      "grad_norm": 9.349382400512695,
+      "learning_rate": 1.1842105263157895e-05,
+      "loss": 0.0213,
+      "step": 870
+    },
+    {
+      "epoch": 2.32,
+      "grad_norm": 11.547332763671875,
+      "learning_rate": 1.1403508771929824e-05,
+      "loss": 0.0526,
+      "step": 880
+    },
+    {
+      "epoch": 2.34,
+      "grad_norm": 4.113059043884277,
+      "learning_rate": 1.0964912280701754e-05,
+      "loss": 0.0187,
+      "step": 890
+    },
+    {
+      "epoch": 2.37,
+      "grad_norm": 3.626723527908325,
+      "learning_rate": 1.0526315789473684e-05,
+      "loss": 0.0234,
+      "step": 900
+    },
+    {
+      "epoch": 2.37,
+      "eval_accuracy": 0.8952569169960475,
+      "eval_f1_macro": 0.8803631688376218,
+      "eval_f1_micro": 0.8952569169960475,
+      "eval_loss": 0.47881799936294556,
+      "eval_runtime": 19.145,
+      "eval_samples_per_second": 79.29,
+      "eval_steps_per_second": 2.507,
+      "step": 900
+    },
+    {
+      "epoch": 2.39,
+      "grad_norm": 11.01301383972168,
+      "learning_rate": 1.0087719298245614e-05,
+      "loss": 0.0655,
+      "step": 910
+    },
+    {
+      "epoch": 2.42,
+      "grad_norm": 11.407805442810059,
+      "learning_rate": 9.649122807017545e-06,
+      "loss": 0.027,
+      "step": 920
+    },
+    {
+      "epoch": 2.45,
+      "grad_norm": 1.768430471420288,
+      "learning_rate": 9.210526315789474e-06,
+      "loss": 0.0518,
+      "step": 930
+    },
+    {
+      "epoch": 2.47,
+      "grad_norm": 20.94036102294922,
+      "learning_rate": 8.771929824561403e-06,
+      "loss": 0.0359,
+      "step": 940
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 1.7965754270553589,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.0145,
+      "step": 950
+    },
+    {
+      "epoch": 2.5,
+      "eval_accuracy": 0.8939393939393939,
+      "eval_f1_macro": 0.877949638322673,
+      "eval_f1_micro": 0.8939393939393939,
+      "eval_loss": 0.49758803844451904,
+      "eval_runtime": 19.1473,
+      "eval_samples_per_second": 79.28,
+      "eval_steps_per_second": 2.507,
+      "step": 950
+    },
+    {
+      "epoch": 2.53,
+      "grad_norm": 0.1458648443222046,
+      "learning_rate": 7.894736842105263e-06,
+      "loss": 0.0425,
+      "step": 960
+    },
+    {
+      "epoch": 2.55,
+      "grad_norm": 11.682854652404785,
+      "learning_rate": 7.456140350877193e-06,
+      "loss": 0.0851,
+      "step": 970
+    },
+    {
+      "epoch": 2.58,
+      "grad_norm": 1.2510184049606323,
+      "learning_rate": 7.017543859649123e-06,
+      "loss": 0.0299,
+      "step": 980
+    },
+    {
+      "epoch": 2.61,
+      "grad_norm": 10.715560913085938,
+      "learning_rate": 6.578947368421053e-06,
+      "loss": 0.0678,
+      "step": 990
+    },
+    {
+      "epoch": 2.63,
+      "grad_norm": 11.586739540100098,
+      "learning_rate": 6.140350877192982e-06,
+      "loss": 0.058,
+      "step": 1000
+    },
+    {
+      "epoch": 2.63,
+      "eval_accuracy": 0.8992094861660079,
+      "eval_f1_macro": 0.8816426015432036,
+      "eval_f1_micro": 0.8992094861660079,
+      "eval_loss": 0.4966810941696167,
+      "eval_runtime": 19.17,
+      "eval_samples_per_second": 79.186,
+      "eval_steps_per_second": 2.504,
+      "step": 1000
+    },
+    {
+      "epoch": 2.66,
+      "grad_norm": 1.5098384618759155,
+      "learning_rate": 5.701754385964912e-06,
+      "loss": 0.0287,
+      "step": 1010
+    },
+    {
+      "epoch": 2.68,
+      "grad_norm": 6.79250955581665,
+      "learning_rate": 5.263157894736842e-06,
+      "loss": 0.0409,
+      "step": 1020
+    },
+    {
+      "epoch": 2.71,
+      "grad_norm": 2.9859375953674316,
+      "learning_rate": 4.824561403508772e-06,
+      "loss": 0.0379,
+      "step": 1030
+    },
+    {
+      "epoch": 2.74,
+      "grad_norm": 14.098068237304688,
+      "learning_rate": 4.3859649122807014e-06,
+      "loss": 0.046,
+      "step": 1040
+    },
+    {
+      "epoch": 2.76,
+      "grad_norm": 10.598516464233398,
+      "learning_rate": 3.9473684210526315e-06,
+      "loss": 0.05,
+      "step": 1050
+    },
+    {
+      "epoch": 2.76,
+      "eval_accuracy": 0.8932806324110671,
+      "eval_f1_macro": 0.8752583058969902,
+      "eval_f1_micro": 0.8932806324110671,
+      "eval_loss": 0.5112892389297485,
+      "eval_runtime": 19.1422,
+      "eval_samples_per_second": 79.301,
+      "eval_steps_per_second": 2.508,
+      "step": 1050
+    },
+    {
+      "epoch": 2.79,
+      "grad_norm": 0.23401279747486115,
+      "learning_rate": 3.5087719298245615e-06,
+      "loss": 0.0655,
+      "step": 1060
+    },
+    {
+      "epoch": 2.82,
+      "grad_norm": 4.760624885559082,
+      "learning_rate": 3.070175438596491e-06,
+      "loss": 0.0659,
+      "step": 1070
+    },
+    {
+      "epoch": 2.84,
+      "grad_norm": 1.6583623886108398,
+      "learning_rate": 2.631578947368421e-06,
+      "loss": 0.0429,
+      "step": 1080
+    },
+    {
+      "epoch": 2.87,
+      "grad_norm": 0.6040318608283997,
+      "learning_rate": 2.1929824561403507e-06,
+      "loss": 0.0168,
+      "step": 1090
+    },
+    {
+      "epoch": 2.89,
+      "grad_norm": 11.433518409729004,
+      "learning_rate": 1.7543859649122807e-06,
+      "loss": 0.0556,
+      "step": 1100
+    },
+    {
+      "epoch": 2.89,
+      "eval_accuracy": 0.8965744400527009,
+      "eval_f1_macro": 0.8803498700160407,
+      "eval_f1_micro": 0.8965744400527009,
+      "eval_loss": 0.502357542514801,
+      "eval_runtime": 19.1404,
+      "eval_samples_per_second": 79.309,
+      "eval_steps_per_second": 2.508,
+      "step": 1100
+    },
+    {
+      "epoch": 2.92,
+      "grad_norm": 17.50218963623047,
+      "learning_rate": 1.3157894736842106e-06,
+      "loss": 0.0795,
+      "step": 1110
+    },
+    {
+      "epoch": 2.95,
+      "grad_norm": 2.289309501647949,
+      "learning_rate": 8.771929824561404e-07,
+      "loss": 0.0646,
+      "step": 1120
+    },
+    {
+      "epoch": 2.97,
+      "grad_norm": 16.876842498779297,
+      "learning_rate": 4.385964912280702e-07,
+      "loss": 0.0431,
+      "step": 1130
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.3004974126815796,
+      "learning_rate": 0.0,
+      "loss": 0.0419,
+      "step": 1140
+    },
+    {
+      "epoch": 3.0,
+      "step": 1140,
+      "total_flos": 1.9041981890455142e+17,
+      "train_loss": 0.48372833394167714,
+      "train_runtime": 1860.5106,
+      "train_samples_per_second": 19.582,
+      "train_steps_per_second": 0.613
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 1140,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 50,
+  "total_flos": 1.9041981890455142e+17,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/amazon_attrprompt/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95b4a5adabb844b5e8347d17e6ffdbb68778d8d70412e41fe9f46cfcf73b127d
+size 6008

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/twitter_disaster/README.md

@@ -0,0 +1,83 @@
+---
+license: other
+library_name: peft
+tags:
+- generated_from_trainer
+base_model: Qwen/Qwen1.5-7B
+metrics:
+- accuracy
+model-index:
+- name: twitter_disaster
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# twitter_disaster
+
+This model is a fine-tuned version of [Qwen/Qwen1.5-7B](https://huggingface.co/Qwen/Qwen1.5-7B) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4902
+- Accuracy: 0.7767
+- F1 Macro: 0.7451
+- F1 Micro: 0.7767
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- total_train_batch_size: 32
+- total_eval_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 3.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:|
+| 0.8422        | 0.18  | 50   | 0.6453          | 0.7178   | 0.6372   | 0.7178   |
+| 0.6082        | 0.37  | 100  | 0.5489          | 0.7472   | 0.7123   | 0.7472   |
+| 0.4305        | 0.55  | 150  | 0.5572          | 0.7252   | 0.5777   | 0.7252   |
+| 0.5021        | 0.74  | 200  | 0.5000          | 0.7721   | 0.7437   | 0.7721   |
+| 0.4715        | 0.92  | 250  | 0.4902          | 0.7767   | 0.7451   | 0.7767   |
+| 0.3937        | 1.1   | 300  | 0.5194          | 0.7601   | 0.7018   | 0.7601   |
+| 0.4219        | 1.29  | 350  | 0.5228          | 0.7665   | 0.7228   | 0.7665   |
+| 0.4315        | 1.47  | 400  | 0.5791          | 0.7555   | 0.6901   | 0.7555   |
+| 0.4134        | 1.65  | 450  | 0.6182          | 0.7390   | 0.7196   | 0.7390   |
+| 0.4173        | 1.84  | 500  | 0.5454          | 0.7638   | 0.7116   | 0.7638   |
+| 0.3278        | 2.02  | 550  | 0.5477          | 0.7721   | 0.7219   | 0.7721   |
+| 0.2641        | 2.21  | 600  | 0.6011          | 0.7528   | 0.7152   | 0.7528   |
+| 0.2256        | 2.39  | 650  | 0.6485          | 0.7601   | 0.6962   | 0.7601   |
+| 0.2544        | 2.57  | 700  | 0.6459          | 0.7629   | 0.7165   | 0.7629   |
+| 0.2839        | 2.76  | 750  | 0.5922          | 0.7656   | 0.7253   | 0.7656   |
+| 0.2634        | 2.94  | 800  | 0.6312          | 0.7638   | 0.7076   | 0.7638   |
+
+
+### Framework versions
+
+- PEFT 0.9.0
+- Transformers 4.39.0.dev0
+- Pytorch 2.2.1+cu121
+- Datasets 2.18.0
+- Tokenizers 0.15.2

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/twitter_disaster/adapter_config.json

@@ -0,0 +1,33 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen1.5-7B",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 256,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "o_proj",
+    "down_proj",
+    "q_proj",
+    "up_proj",
+    "k_proj",
+    "gate_proj"
+  ],
+  "task_type": "SEQ_CLS",
+  "use_dora": false,
+  "use_rslora": false
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/twitter_disaster/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:301692a0eb07d9c0bd5e5640812ddc1e8c3bc1f6d8345dd38a10122ec5be8f96
+size 1882088146

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/twitter_disaster/added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/twitter_disaster/all_results.json

@@ -0,0 +1,23 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.7766544117647058,
+    "eval_f1_macro": 0.7450627015924902,
+    "eval_f1_micro": 0.7766544117647058,
+    "eval_loss": 0.4901912808418274,
+    "eval_runtime": 13.0776,
+    "eval_samples": 1088,
+    "eval_samples_per_second": 83.196,
+    "eval_steps_per_second": 2.6,
+    "test_accuracy": 0.7766544117647058,
+    "test_f1_macro": 0.7415457166235069,
+    "test_f1_micro": 0.7766544117647058,
+    "test_loss": 0.4903779923915863,
+    "test_runtime": 13.1647,
+    "test_samples_per_second": 82.645,
+    "test_steps_per_second": 2.583,
+    "train_loss": 0.4493609409706265,
+    "train_runtime": 1238.3624,
+    "train_samples": 8700,
+    "train_samples_per_second": 21.076,
+    "train_steps_per_second": 0.659
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/twitter_disaster/checkpoint-250/README.md

@@ -0,0 +1,202 @@
+---
+library_name: peft
+base_model: Qwen/Qwen1.5-7B
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.9.0

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/twitter_disaster/checkpoint-250/adapter_config.json

@@ -0,0 +1,33 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen1.5-7B",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 256,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "o_proj",
+    "down_proj",
+    "q_proj",
+    "up_proj",
+    "k_proj",
+    "gate_proj"
+  ],
+  "task_type": "SEQ_CLS",
+  "use_dora": false,
+  "use_rslora": false
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/twitter_disaster/checkpoint-250/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:301692a0eb07d9c0bd5e5640812ddc1e8c3bc1f6d8345dd38a10122ec5be8f96
+size 1882088146

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/twitter_disaster/checkpoint-250/added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/twitter_disaster/checkpoint-250/global_step250/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6619ff7cd9ef8e4cb0086da16a12d210a09c4bfd3f22c6822735f7da661302c5
+size 1918971376

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/twitter_disaster/checkpoint-250/global_step250/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b90415212486928841f9fbef29085dc4403133bc6be77ae722c3ecb8e3a384a
+size 1918971504

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/twitter_disaster/checkpoint-250/global_step250/mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed9aa0e20325f480fc1b8f4a6ec13b27d874d7becdad95231dedd690faab474a
+size 639974444

LoRA/Qwen/Qwen1.5_7B_LoRA_MAdAiLab/twitter_disaster/checkpoint-250/latest

@@ -0,0 +1 @@
+global_step250