Upload folder using huggingface_hub

README.md

@@ -0,0 +1,277 @@
+---
+language: fr
+license: mit
+tags:
+- deberta-v2
+- token-classification
+base_model: almanach/camembertav2-base
+datasets:
+- FTB-NER
+metrics:
+- f1
+pipeline_tag: token-classification
+library_name: transformers
+model-index:
+- name: almanach/camembertav2-base-ftb-ner
+  results:
+  - task:
+      type: token-classification
+      name: Named Entity Recognition (NER)
+    dataset:
+      type: ftb-ner
+      name: French Treebank Named Entity Recognition
+    metrics:
+    - name: f1
+      type: f1
+      value: 0.93966
+      verified: false
+---
+
+# Model Card for almanach/camembertav2-base-ftb-ner
+
+almanach/camembertav2-base-ftb-ner is a deberta-v2 model for token classification. It is trained on the FTB-NER dataset for the task of Named Entity Recognition (NER). The model achieves an f1 score of 0.93966 on the FTB-NER dataset.
+
+The model is part of the almanach/camembertav2-base family of model finetunes.
+
+## Model Details
+
+### Model Description
+
+- **Developed by:** Wissam Antoun (Phd Student at Almanach, Inria-Paris)
+- **Model type:** deberta-v2
+- **Language(s) (NLP):** French
+- **License:** MIT
+- **Finetuned from model [optional]:** almanach/camembertav2-base
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** https://github.com/WissamAntoun/camemberta
+- **Paper:** https://arxiv.org/abs/2411.08868
+
+## Uses
+
+The model can be used for token classification tasks in French for Named Entity Recognition (NER).
+
+## Bias, Risks, and Limitations
+
+The model may exhibit biases based on the training data. The model may not generalize well to other datasets or tasks. The model may also have limitations in terms of the data it was trained on.
+
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+```python
+from transformers import AutoTokenizer, AutoModelForTokenClassification, pipeline
+
+model = AutoModelForTokenClassification.from_pretrained("almanach/camembertav2-base-ftb-ner")
+tokenizer = AutoTokenizer.from_pretrained("almanach/camembertav2-base-ftb-ner")
+
+classifier = pipeline("token-classification", model=model, tokenizer=tokenizer)
+
+classifier("Votre texte ici")
+```
+
+
+## Training Details
+
+### Training Data
+
+The model is trained on the FTB-NER dataset.
+
+- Dataset Name: FTB-NER
+- Dataset Size:
+    - Train: 9881
+    - Dev: 1235
+    - Test: 1235
+
+
+### Training Procedure
+
+Model trained with the run_ner.py script from the huggingface repository.
+
+
+
+#### Training Hyperparameters
+
+```yml
+accelerator_config: '{''split_batches'': False, ''dispatch_batches'': None, ''even_batches'':
+  True, ''use_seedable_sampler'': True, ''non_blocking'': False, ''gradient_accumulation_kwargs'':
+  None}'
+adafactor: false
+adam_beta1: 0.9
+adam_beta2: 0.999
+adam_epsilon: 1.0e-08
+auto_find_batch_size: false
+base_model: camembertv2
+base_model_name: camembertav2-base-bf16-p2-17000
+batch_eval_metrics: false
+bf16: false
+bf16_full_eval: false
+data_seed: 42.0
+dataloader_drop_last: false
+dataloader_num_workers: 0
+dataloader_persistent_workers: false
+dataloader_pin_memory: true
+dataloader_prefetch_factor: .nan
+ddp_backend: .nan
+ddp_broadcast_buffers: .nan
+ddp_bucket_cap_mb: .nan
+ddp_find_unused_parameters: .nan
+ddp_timeout: 1800
+debug: '[]'
+deepspeed: .nan
+disable_tqdm: false
+dispatch_batches: .nan
+do_eval: true
+do_predict: false
+do_train: true
+epoch: 8.0
+eval_accumulation_steps: 4
+eval_accuracy: 0.9941108798071656
+eval_delay: 0
+eval_do_concat_batches: true
+eval_f1: 0.939655172413793
+eval_loss: 0.0354763492941856
+eval_on_start: false
+eval_precision: 0.93671875
+eval_recall: 0.9426100628930818
+eval_runtime: 3.729
+eval_samples: 1235.0
+eval_samples_per_second: 331.184
+eval_steps: .nan
+eval_steps_per_second: 41.566
+eval_strategy: epoch
+eval_use_gather_object: false
+evaluation_strategy: epoch
+fp16: false
+fp16_backend: auto
+fp16_full_eval: false
+fp16_opt_level: O1
+fsdp: '[]'
+fsdp_config: '{''min_num_params'': 0, ''xla'': False, ''xla_fsdp_v2'': False, ''xla_fsdp_grad_ckpt'':
+  False}'
+fsdp_min_num_params: 0
+fsdp_transformer_layer_cls_to_wrap: .nan
+full_determinism: false
+gradient_accumulation_steps: 2
+gradient_checkpointing: false
+gradient_checkpointing_kwargs: .nan
+greater_is_better: true
+group_by_length: false
+half_precision_backend: auto
+hub_always_push: false
+hub_model_id: .nan
+hub_private_repo: false
+hub_strategy: every_save
+hub_token: <HUB_TOKEN>
+ignore_data_skip: false
+include_inputs_for_metrics: false
+include_num_input_tokens_seen: false
+include_tokens_per_second: false
+jit_mode_eval: false
+label_names: .nan
+label_smoothing_factor: 0.0
+learning_rate: 5.000000000000001e-05
+length_column_name: length
+load_best_model_at_end: true
+local_rank: 0
+log_level: debug
+log_level_replica: warning
+log_on_each_node: true
+logging_dir: /scratch/camembertv2/runs/results/ftb_ner/camembertav2-base-bf16-p2-17000/max_seq_length-192-gradient_accumulation_steps-2-precision-fp32-learning_rate-5.000000000000001e-05-epochs-8-lr_scheduler-linear-warmup_steps-0.1/SEED-42/logs
+logging_first_step: false
+logging_nan_inf_filter: true
+logging_steps: 100
+logging_strategy: steps
+lr_scheduler_kwargs: '{}'
+lr_scheduler_type: linear
+max_grad_norm: 1.0
+max_steps: -1
+metric_for_best_model: f1
+mp_parameters: .nan
+name: camembertv2/runs/results/ftb_ner/camembertav2-base-bf16-p2-17000/max_seq_length-192-gradient_accumulation_steps-2-precision-fp32-learning_rate-5.000000000000001e-05-epochs-8-lr_scheduler-linear-warmup_steps-0.1
+neftune_noise_alpha: .nan
+no_cuda: false
+num_train_epochs: 8.0
+optim: adamw_torch
+optim_args: .nan
+optim_target_modules: .nan
+output_dir: /scratch/camembertv2/runs/results/ftb_ner/camembertav2-base-bf16-p2-17000/max_seq_length-192-gradient_accumulation_steps-2-precision-fp32-learning_rate-5.000000000000001e-05-epochs-8-lr_scheduler-linear-warmup_steps-0.1/SEED-42
+overwrite_output_dir: false
+past_index: -1
+per_device_eval_batch_size: 8
+per_device_train_batch_size: 8
+per_gpu_eval_batch_size: .nan
+per_gpu_train_batch_size: .nan
+prediction_loss_only: false
+push_to_hub: false
+push_to_hub_model_id: .nan
+push_to_hub_organization: .nan
+push_to_hub_token: <PUSH_TO_HUB_TOKEN>
+ray_scope: last
+remove_unused_columns: true
+report_to: '[''tensorboard'']'
+restore_callback_states_from_checkpoint: false
+resume_from_checkpoint: .nan
+run_name: /scratch/camembertv2/runs/results/ftb_ner/camembertav2-base-bf16-p2-17000/max_seq_length-192-gradient_accumulation_steps-2-precision-fp32-learning_rate-5.000000000000001e-05-epochs-8-lr_scheduler-linear-warmup_steps-0.1/SEED-42
+save_on_each_node: false
+save_only_model: false
+save_safetensors: true
+save_steps: 500
+save_strategy: epoch
+save_total_limit: .nan
+seed: 42
+skip_memory_metrics: true
+split_batches: .nan
+tf32: .nan
+torch_compile: true
+torch_compile_backend: inductor
+torch_compile_mode: .nan
+torch_empty_cache_steps: .nan
+torchdynamo: .nan
+total_flos: 2834249641269840.0
+tpu_metrics_debug: false
+tpu_num_cores: .nan
+train_loss: 0.0513342066417812
+train_runtime: 849.5258
+train_samples: 9881
+train_samples_per_second: 93.05
+train_steps_per_second: 5.82
+use_cpu: false
+use_ipex: false
+use_legacy_prediction_loop: false
+use_mps_device: false
+warmup_ratio: 0.1
+warmup_steps: 0
+weight_decay: 0.0
+
+```
+
+#### Results
+
+**F1-Score:** 0.93966
+
+## Technical Specifications
+
+### Model Architecture and Objective
+
+deberta-v2 for token classification.
+
+## Citation
+
+**BibTeX:**
+
+```bibtex
+@misc{antoun2024camembert20smarterfrench,
+      title={CamemBERT 2.0: A Smarter French Language Model Aged to Perfection},
+      author={Wissam Antoun and Francis Kulumba and Rian Touchent and Éric de la Clergerie and Benoît Sagot and Djamé Seddah},
+      year={2024},
+      eprint={2411.08868},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL},
+      url={https://arxiv.org/abs/2411.08868},
+}
+```

all_results.json

@@ -0,0 +1,18 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.9941108798071655,
+    "eval_f1": 0.939655172413793,
+    "eval_loss": 0.03547634929418564,
+    "eval_precision": 0.93671875,
+    "eval_recall": 0.9426100628930818,
+    "eval_runtime": 3.729,
+    "eval_samples": 1235,
+    "eval_samples_per_second": 331.184,
+    "eval_steps_per_second": 41.566,
+    "total_flos": 2834249641269840.0,
+    "train_loss": 0.05133420664178129,
+    "train_runtime": 849.5258,
+    "train_samples": 9881,
+    "train_samples_per_second": 93.05,
+    "train_steps_per_second": 5.82
+}

config.json

@@ -0,0 +1,76 @@
+{
+  "_name_or_path": "/scratch/camembertv2/runs/models/camembertav2-base-bf16/post/ckpt-p2-17000/pt/discriminator/",
+  "architectures": [
+    "DebertaV2ForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 1,
+  "conv_act": "gelu",
+  "conv_kernel_size": 0,
+  "embedding_size": 768,
+  "eos_token_id": 2,
+  "finetuning_task": "ner",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "B-Company",
+    "1": "B-FictionCharacter",
+    "2": "B-Location",
+    "3": "B-Organization",
+    "4": "B-Person",
+    "5": "B-POI",
+    "6": "B-Product",
+    "7": "I-Company",
+    "8": "I-FictionCharacter",
+    "9": "I-Location",
+    "10": "I-Organization",
+    "11": "I-Person",
+    "12": "I-POI",
+    "13": "I-Product",
+    "14": "O"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "B-Company": 0,
+    "B-FictionCharacter": 1,
+    "B-Location": 2,
+    "B-Organization": 3,
+    "B-POI": 5,
+    "B-Person": 4,
+    "B-Product": 6,
+    "I-Company": 7,
+    "I-FictionCharacter": 8,
+    "I-Location": 9,
+    "I-Organization": 10,
+    "I-POI": 12,
+    "I-Person": 11,
+    "I-Product": 13,
+    "O": 14
+  },
+  "layer_norm_eps": 1e-07,
+  "max_position_embeddings": 1024,
+  "max_relative_positions": -1,
+  "model_name": "camembertav2-base-bf16",
+  "model_type": "deberta-v2",
+  "norm_rel_ebd": "layer_norm",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 768,
+  "pos_att_type": [
+    "p2c",
+    "c2p"
+  ],
+  "position_biased_input": false,
+  "position_buckets": 256,
+  "relative_attention": true,
+  "share_att_key": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.44.2",
+  "type_vocab_size": 0,
+  "vocab_size": 32768
+}

eval_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.9941108798071655,
+    "eval_f1": 0.939655172413793,
+    "eval_loss": 0.03547634929418564,
+    "eval_precision": 0.93671875,
+    "eval_recall": 0.9426100628930818,
+    "eval_runtime": 3.729,
+    "eval_samples": 1235,
+    "eval_samples_per_second": 331.184,
+    "eval_steps_per_second": 41.566
+}

logs/events.out.tfevents.1724615707.nefgpu51.118010.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a609b07ce07b9b92609080d6d2122e0d8192a1ef9f7c8bee421fd53849c447a9
+size 21110

logs/events.out.tfevents.1724616561.nefgpu51.118010.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cbc296350b9f30d51ec6a61703c1bb764ebdeca2ac2043fc66796c3b8d5107f7
+size 512

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe063cb24c9a33f5eb6c216ae732f1c1b7fb4d49f9fee90597a72e14088f073a
+size 442536820

special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,57 @@
+{
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "[CLS]",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "eos_token": "[SEP]",
+  "errors": "replace",
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "[UNK]"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "total_flos": 2834249641269840.0,
+    "train_loss": 0.05133420664178129,
+    "train_runtime": 849.5258,
+    "train_samples": 9881,
+    "train_samples_per_second": 93.05,
+    "train_steps_per_second": 5.82
+}

trainer_state.json

@@ -0,0 +1,481 @@
+{
+  "best_metric": 0.939655172413793,
+  "best_model_checkpoint": "/scratch/camembertv2/runs/results/ftb_ner/camembertav2-base-bf16-p2-17000/max_seq_length-192-gradient_accumulation_steps-2-precision-fp32-learning_rate-5.000000000000001e-05-epochs-8-lr_scheduler-linear-warmup_steps-0.1/SEED-42/checkpoint-4944",
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 4944,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.16181229773462782,
+      "grad_norm": 0.7081506848335266,
+      "learning_rate": 1.0101010101010103e-05,
+      "loss": 1.713,
+      "step": 100
+    },
+    {
+      "epoch": 0.32362459546925565,
+      "grad_norm": 0.6198400855064392,
+      "learning_rate": 2.0202020202020206e-05,
+      "loss": 0.1665,
+      "step": 200
+    },
+    {
+      "epoch": 0.4854368932038835,
+      "grad_norm": 0.813007652759552,
+      "learning_rate": 3.030303030303031e-05,
+      "loss": 0.0816,
+      "step": 300
+    },
+    {
+      "epoch": 0.6472491909385113,
+      "grad_norm": 0.9836126565933228,
+      "learning_rate": 4.040404040404041e-05,
+      "loss": 0.0487,
+      "step": 400
+    },
+    {
+      "epoch": 0.8090614886731392,
+      "grad_norm": 0.2043960988521576,
+      "learning_rate": 4.9943807597212865e-05,
+      "loss": 0.0415,
+      "step": 500
+    },
+    {
+      "epoch": 0.970873786407767,
+      "grad_norm": 0.6549485921859741,
+      "learning_rate": 4.8819959541470004e-05,
+      "loss": 0.0435,
+      "step": 600
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9896461049632957,
+      "eval_f1": 0.8642940946421588,
+      "eval_loss": 0.04151414707303047,
+      "eval_precision": 0.8599221789883269,
+      "eval_recall": 0.8687106918238994,
+      "eval_runtime": 4.2561,
+      "eval_samples_per_second": 290.171,
+      "eval_steps_per_second": 36.418,
+      "step": 618
+    },
+    {
+      "epoch": 1.132686084142395,
+      "grad_norm": 0.905681848526001,
+      "learning_rate": 4.7696111485727136e-05,
+      "loss": 0.0221,
+      "step": 700
+    },
+    {
+      "epoch": 1.2944983818770226,
+      "grad_norm": 0.4660435616970062,
+      "learning_rate": 4.6572263429984275e-05,
+      "loss": 0.029,
+      "step": 800
+    },
+    {
+      "epoch": 1.4563106796116505,
+      "grad_norm": 0.4658781886100769,
+      "learning_rate": 4.544841537424141e-05,
+      "loss": 0.0265,
+      "step": 900
+    },
+    {
+      "epoch": 1.6181229773462782,
+      "grad_norm": 1.3315564393997192,
+      "learning_rate": 4.4324567318498546e-05,
+      "loss": 0.022,
+      "step": 1000
+    },
+    {
+      "epoch": 1.779935275080906,
+      "grad_norm": 0.08498021215200424,
+      "learning_rate": 4.3200719262755685e-05,
+      "loss": 0.025,
+      "step": 1100
+    },
+    {
+      "epoch": 1.941747572815534,
+      "grad_norm": 0.4140647351741791,
+      "learning_rate": 4.207687120701282e-05,
+      "loss": 0.0386,
+      "step": 1200
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9920017530404295,
+      "eval_f1": 0.8975457732761979,
+      "eval_loss": 0.035175956785678864,
+      "eval_precision": 0.8895752895752895,
+      "eval_recall": 0.9056603773584906,
+      "eval_runtime": 3.7654,
+      "eval_samples_per_second": 327.989,
+      "eval_steps_per_second": 41.165,
+      "step": 1236
+    },
+    {
+      "epoch": 2.103559870550162,
+      "grad_norm": 0.43421319127082825,
+      "learning_rate": 4.0953023151269956e-05,
+      "loss": 0.0249,
+      "step": 1300
+    },
+    {
+      "epoch": 2.26537216828479,
+      "grad_norm": 0.2701377272605896,
+      "learning_rate": 3.9829175095527095e-05,
+      "loss": 0.0191,
+      "step": 1400
+    },
+    {
+      "epoch": 2.4271844660194173,
+      "grad_norm": 0.11492807418107986,
+      "learning_rate": 3.870532703978423e-05,
+      "loss": 0.0136,
+      "step": 1500
+    },
+    {
+      "epoch": 2.588996763754045,
+      "grad_norm": 0.582848072052002,
+      "learning_rate": 3.758147898404136e-05,
+      "loss": 0.0258,
+      "step": 1600
+    },
+    {
+      "epoch": 2.750809061488673,
+      "grad_norm": 0.2515222132205963,
+      "learning_rate": 3.6457630928298505e-05,
+      "loss": 0.0169,
+      "step": 1700
+    },
+    {
+      "epoch": 2.912621359223301,
+      "grad_norm": 0.36368367075920105,
+      "learning_rate": 3.533378287255564e-05,
+      "loss": 0.0123,
+      "step": 1800
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9929330557685987,
+      "eval_f1": 0.9168309026409145,
+      "eval_loss": 0.030728859826922417,
+      "eval_precision": 0.9193675889328063,
+      "eval_recall": 0.914308176100629,
+      "eval_runtime": 3.7477,
+      "eval_samples_per_second": 329.532,
+      "eval_steps_per_second": 41.358,
+      "step": 1854
+    },
+    {
+      "epoch": 3.074433656957929,
+      "grad_norm": 0.3048587739467621,
+      "learning_rate": 3.420993481681277e-05,
+      "loss": 0.0128,
+      "step": 1900
+    },
+    {
+      "epoch": 3.236245954692557,
+      "grad_norm": 0.01976470835506916,
+      "learning_rate": 3.308608676106991e-05,
+      "loss": 0.0145,
+      "step": 2000
+    },
+    {
+      "epoch": 3.3980582524271843,
+      "grad_norm": 0.17214246094226837,
+      "learning_rate": 3.196223870532705e-05,
+      "loss": 0.0092,
+      "step": 2100
+    },
+    {
+      "epoch": 3.559870550161812,
+      "grad_norm": 0.2264121174812317,
+      "learning_rate": 3.083839064958418e-05,
+      "loss": 0.0111,
+      "step": 2200
+    },
+    {
+      "epoch": 3.72168284789644,
+      "grad_norm": 0.13608065247535706,
+      "learning_rate": 2.971454259384132e-05,
+      "loss": 0.0124,
+      "step": 2300
+    },
+    {
+      "epoch": 3.883495145631068,
+      "grad_norm": 0.027897851541638374,
+      "learning_rate": 2.8590694538098453e-05,
+      "loss": 0.0136,
+      "step": 2400
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9914265366495015,
+      "eval_f1": 0.9107497024990083,
+      "eval_loss": 0.03449518606066704,
+      "eval_precision": 0.9191353082465973,
+      "eval_recall": 0.9025157232704403,
+      "eval_runtime": 3.7618,
+      "eval_samples_per_second": 328.3,
+      "eval_steps_per_second": 41.204,
+      "step": 2472
+    },
+    {
+      "epoch": 4.0453074433656955,
+      "grad_norm": 0.18220163881778717,
+      "learning_rate": 2.746684648235559e-05,
+      "loss": 0.0088,
+      "step": 2500
+    },
+    {
+      "epoch": 4.207119741100324,
+      "grad_norm": 0.016186678782105446,
+      "learning_rate": 2.6342998426612728e-05,
+      "loss": 0.0087,
+      "step": 2600
+    },
+    {
+      "epoch": 4.368932038834951,
+      "grad_norm": 0.18213553726673126,
+      "learning_rate": 2.5219150370869863e-05,
+      "loss": 0.0051,
+      "step": 2700
+    },
+    {
+      "epoch": 4.53074433656958,
+      "grad_norm": 0.012032161466777325,
+      "learning_rate": 2.4095302315127e-05,
+      "loss": 0.0081,
+      "step": 2800
+    },
+    {
+      "epoch": 4.692556634304207,
+      "grad_norm": 0.0591534860432148,
+      "learning_rate": 2.2971454259384134e-05,
+      "loss": 0.0042,
+      "step": 2900
+    },
+    {
+      "epoch": 4.854368932038835,
+      "grad_norm": 0.3236534893512726,
+      "learning_rate": 2.1847606203641273e-05,
+      "loss": 0.009,
+      "step": 3000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9935904459296593,
+      "eval_f1": 0.9192352711666016,
+      "eval_loss": 0.029976682737469673,
+      "eval_precision": 0.9124709527498064,
+      "eval_recall": 0.9261006289308176,
+      "eval_runtime": 3.7424,
+      "eval_samples_per_second": 330.006,
+      "eval_steps_per_second": 41.418,
+      "step": 3090
+    },
+    {
+      "epoch": 5.016181229773463,
+      "grad_norm": 0.006900902837514877,
+      "learning_rate": 2.072375814789841e-05,
+      "loss": 0.0089,
+      "step": 3100
+    },
+    {
+      "epoch": 5.17799352750809,
+      "grad_norm": 0.10268358141183853,
+      "learning_rate": 1.9599910092155544e-05,
+      "loss": 0.003,
+      "step": 3200
+    },
+    {
+      "epoch": 5.339805825242719,
+      "grad_norm": 0.008188015781342983,
+      "learning_rate": 1.847606203641268e-05,
+      "loss": 0.003,
+      "step": 3300
+    },
+    {
+      "epoch": 5.501618122977346,
+      "grad_norm": 0.159342423081398,
+      "learning_rate": 1.735221398066982e-05,
+      "loss": 0.0027,
+      "step": 3400
+    },
+    {
+      "epoch": 5.663430420711974,
+      "grad_norm": 0.4539357125759125,
+      "learning_rate": 1.622836592492695e-05,
+      "loss": 0.0042,
+      "step": 3500
+    },
+    {
+      "epoch": 5.825242718446602,
+      "grad_norm": 0.038250233978033066,
+      "learning_rate": 1.510451786918409e-05,
+      "loss": 0.0039,
+      "step": 3600
+    },
+    {
+      "epoch": 5.9870550161812295,
+      "grad_norm": 0.004649197217077017,
+      "learning_rate": 1.3980669813441227e-05,
+      "loss": 0.0038,
+      "step": 3700
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9935630546729484,
+      "eval_f1": 0.9324960753532182,
+      "eval_loss": 0.033860765397548676,
+      "eval_precision": 0.9310344827586207,
+      "eval_recall": 0.9339622641509434,
+      "eval_runtime": 3.7552,
+      "eval_samples_per_second": 328.879,
+      "eval_steps_per_second": 41.276,
+      "step": 3708
+    },
+    {
+      "epoch": 6.148867313915858,
+      "grad_norm": 0.20126421749591827,
+      "learning_rate": 1.285682175769836e-05,
+      "loss": 0.0016,
+      "step": 3800
+    },
+    {
+      "epoch": 6.310679611650485,
+      "grad_norm": 0.006144899874925613,
+      "learning_rate": 1.1732973701955498e-05,
+      "loss": 0.0024,
+      "step": 3900
+    },
+    {
+      "epoch": 6.472491909385114,
+      "grad_norm": 1.5297178030014038,
+      "learning_rate": 1.0609125646212633e-05,
+      "loss": 0.0022,
+      "step": 4000
+    },
+    {
+      "epoch": 6.634304207119741,
+      "grad_norm": 0.04670717939734459,
+      "learning_rate": 9.48527759046977e-06,
+      "loss": 0.002,
+      "step": 4100
+    },
+    {
+      "epoch": 6.796116504854369,
+      "grad_norm": 0.028466541320085526,
+      "learning_rate": 8.361429534726907e-06,
+      "loss": 0.0018,
+      "step": 4200
+    },
+    {
+      "epoch": 6.957928802588997,
+      "grad_norm": 0.09033439308404922,
+      "learning_rate": 7.237581478984042e-06,
+      "loss": 0.0019,
+      "step": 4300
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.9936726196997918,
+      "eval_f1": 0.9328125,
+      "eval_loss": 0.03374877944588661,
+      "eval_precision": 0.9270186335403726,
+      "eval_recall": 0.9386792452830188,
+      "eval_runtime": 3.7709,
+      "eval_samples_per_second": 327.506,
+      "eval_steps_per_second": 41.104,
+      "step": 4326
+    },
+    {
+      "epoch": 7.119741100323624,
+      "grad_norm": 0.020341284573078156,
+      "learning_rate": 6.113733423241179e-06,
+      "loss": 0.0017,
+      "step": 4400
+    },
+    {
+      "epoch": 7.281553398058253,
+      "grad_norm": 0.026211928576231003,
+      "learning_rate": 4.989885367498316e-06,
+      "loss": 0.0009,
+      "step": 4500
+    },
+    {
+      "epoch": 7.44336569579288,
+      "grad_norm": 0.0029867186676710844,
+      "learning_rate": 3.866037311755451e-06,
+      "loss": 0.0014,
+      "step": 4600
+    },
+    {
+      "epoch": 7.605177993527509,
+      "grad_norm": 0.014907300472259521,
+      "learning_rate": 2.742189256012588e-06,
+      "loss": 0.0014,
+      "step": 4700
+    },
+    {
+      "epoch": 7.766990291262136,
+      "grad_norm": 0.12841272354125977,
+      "learning_rate": 1.6183412002697239e-06,
+      "loss": 0.0013,
+      "step": 4800
+    },
+    {
+      "epoch": 7.9288025889967635,
+      "grad_norm": 0.0018621513154357672,
+      "learning_rate": 4.9449314452686e-07,
+      "loss": 0.0013,
+      "step": 4900
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.9941108798071655,
+      "eval_f1": 0.939655172413793,
+      "eval_loss": 0.03547634929418564,
+      "eval_precision": 0.93671875,
+      "eval_recall": 0.9426100628930818,
+      "eval_runtime": 3.7628,
+      "eval_samples_per_second": 328.211,
+      "eval_steps_per_second": 41.192,
+      "step": 4944
+    },
+    {
+      "epoch": 8.0,
+      "step": 4944,
+      "total_flos": 2834249641269840.0,
+      "train_loss": 0.05133420664178129,
+      "train_runtime": 849.5258,
+      "train_samples_per_second": 93.05,
+      "train_steps_per_second": 5.82
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 4944,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2834249641269840.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41dcd20c450f3eec9cc4fb37650c7eecf9a81c19de46d3a3ddd51572b43a373a
+size 5624