Upload folder using huggingface_hub

README.md

@@ -0,0 +1,277 @@
+---
+language: fr
+license: mit
+tags:
+- roberta
+- token-classification
+base_model: almanach/camembertv2-base
+datasets:
+- FTB-NER
+metrics:
+- f1
+pipeline_tag: token-classification
+library_name: transformers
+model-index:
+- name: almanach/camembertv2-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: 0.93548
+      verified: false
+---
+
+# Model Card for almanach/camembertv2-base-ftb-ner
+
+almanach/camembertv2-base-ftb-ner is a roberta 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 0.93548 on the FTB-NER dataset.
+
+The model is part of the almanach/camembertv2-base family of model finetunes.
+
+## Model Details
+
+### Model Description
+
+- **Developed by:** Wissam Antoun (Phd Student at Almanach, Inria-Paris)
+- **Model type:** roberta
+- **Language(s) (NLP):** French
+- **License:** MIT
+- **Finetuned from model [optional]:** almanach/camembertv2-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/camembertv2-base-ftb-ner")
+tokenizer = AutoTokenizer.from_pretrained("almanach/camembertv2-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: camembertv2-base-bf16-p2-17000
+batch_eval_metrics: false
+bf16: false
+bf16_full_eval: false
+data_seed: 1337.0
+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: 0.9937000109565028
+eval_delay: 0
+eval_do_concat_batches: true
+eval_f1: 0.935483870967742
+eval_loss: 0.0347304567694664
+eval_on_start: false
+eval_precision: 0.9362204724409448
+eval_recall: 0.934748427672956
+eval_runtime: 2.7702
+eval_samples: 1235.0
+eval_samples_per_second: 445.821
+eval_steps: .nan
+eval_steps_per_second: 55.953
+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/camembertv2-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-1337/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/camembertv2-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/camembertv2-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-1337
+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/camembertv2-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-1337
+save_on_each_node: false
+save_only_model: false
+save_safetensors: true
+save_steps: 500
+save_strategy: epoch
+save_total_limit: .nan
+seed: 1337
+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: 2833132740217920.0
+tpu_metrics_debug: false
+tpu_num_cores: .nan
+train_loss: 0.0880794880495777
+train_runtime: 679.3683
+train_samples: 9881
+train_samples_per_second: 116.355
+train_steps_per_second: 7.277
+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:** 0.93548
+
+## Technical Specifications
+
+### Model Architecture and Objective
+
+roberta 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": 0.9937000109565027,
+    "eval_f1": 0.935483870967742,
+    "eval_loss": 0.0347304567694664,
+    "eval_precision": 0.9362204724409449,
+    "eval_recall": 0.934748427672956,
+    "eval_runtime": 2.7702,
+    "eval_samples": 1235,
+    "eval_samples_per_second": 445.821,
+    "eval_steps_per_second": 55.953,
+    "total_flos": 2833132740217920.0,
+    "train_loss": 0.08807948804957774,
+    "train_runtime": 679.3683,
+    "train_samples": 9881,
+    "train_samples_per_second": 116.355,
+    "train_steps_per_second": 7.277
+}

config.json

@@ -0,0 +1,65 @@
+{
+  "_name_or_path": "/scratch/camembertv2/runs/models/camembertv2-base-bf16/post/ckpt-p2-17000/pt/",
+  "architectures": [
+    "RobertaForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 1,
+  "classifier_dropout": null,
+  "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": 1025,
+  "model_name": "camembertv2-base-bf16",
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_biased_input": true,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.44.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 32768
+}

eval_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.9937000109565027,
+    "eval_f1": 0.935483870967742,
+    "eval_loss": 0.0347304567694664,
+    "eval_precision": 0.9362204724409449,
+    "eval_recall": 0.934748427672956,
+    "eval_runtime": 2.7702,
+    "eval_samples": 1235,
+    "eval_samples_per_second": 445.821,
+    "eval_steps_per_second": 55.953
+}

logs/events.out.tfevents.1724620471.nefgpu51.144270.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05e0dfd1d280d26e46351f866ba2d2f08b3454e69b9cb8753372402a245240e7
+size 20871

logs/events.out.tfevents.1724621154.nefgpu51.144270.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13c7b317b552a6e5428a9115dc9a4dcacaee232a5f8b1063e30f1f833ecc63c9
+size 512

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed6da0d29a943fc0383213e4349a636a6241359582e6f0d94e9a3f500ab615b8
+size 444109236

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": 2833132740217920.0,
+    "train_loss": 0.08807948804957774,
+    "train_runtime": 679.3683,
+    "train_samples": 9881,
+    "train_samples_per_second": 116.355,
+    "train_steps_per_second": 7.277
+}

trainer_state.json

@@ -0,0 +1,481 @@
+{
+  "best_metric": 0.935483870967742,
+  "best_model_checkpoint": "/scratch/camembertv2/runs/results/ftb_ner/camembertv2-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-1337/checkpoint-4326",
+  "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": 9.89955997467041,
+      "learning_rate": 1.0101010101010103e-05,
+      "loss": 1.8738,
+      "step": 100
+    },
+    {
+      "epoch": 0.32362459546925565,
+      "grad_norm": 2.3764805793762207,
+      "learning_rate": 2.0202020202020206e-05,
+      "loss": 0.6979,
+      "step": 200
+    },
+    {
+      "epoch": 0.4854368932038835,
+      "grad_norm": 1.3664543628692627,
+      "learning_rate": 3.030303030303031e-05,
+      "loss": 0.5111,
+      "step": 300
+    },
+    {
+      "epoch": 0.6472491909385113,
+      "grad_norm": 0.6372264623641968,
+      "learning_rate": 4.040404040404041e-05,
+      "loss": 0.2666,
+      "step": 400
+    },
+    {
+      "epoch": 0.8090614886731392,
+      "grad_norm": 0.5098221302032471,
+      "learning_rate": 4.9943807597212865e-05,
+      "loss": 0.1199,
+      "step": 500
+    },
+    {
+      "epoch": 0.970873786407767,
+      "grad_norm": 0.5974541902542114,
+      "learning_rate": 4.8819959541470004e-05,
+      "loss": 0.0775,
+      "step": 600
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9852635038895584,
+      "eval_f1": 0.7820512820512822,
+      "eval_loss": 0.0750068947672844,
+      "eval_precision": 0.7514492753623189,
+      "eval_recall": 0.815251572327044,
+      "eval_runtime": 3.2799,
+      "eval_samples_per_second": 376.537,
+      "eval_steps_per_second": 47.258,
+      "step": 618
+    },
+    {
+      "epoch": 1.132686084142395,
+      "grad_norm": 0.15989889204502106,
+      "learning_rate": 4.7696111485727136e-05,
+      "loss": 0.0648,
+      "step": 700
+    },
+    {
+      "epoch": 1.2944983818770226,
+      "grad_norm": 0.28292131423950195,
+      "learning_rate": 4.6572263429984275e-05,
+      "loss": 0.0555,
+      "step": 800
+    },
+    {
+      "epoch": 1.4563106796116505,
+      "grad_norm": 0.09367953985929489,
+      "learning_rate": 4.544841537424141e-05,
+      "loss": 0.0485,
+      "step": 900
+    },
+    {
+      "epoch": 1.6181229773462782,
+      "grad_norm": 0.3826428949832916,
+      "learning_rate": 4.4324567318498546e-05,
+      "loss": 0.0401,
+      "step": 1000
+    },
+    {
+      "epoch": 1.779935275080906,
+      "grad_norm": 0.18068315088748932,
+      "learning_rate": 4.3200719262755685e-05,
+      "loss": 0.0369,
+      "step": 1100
+    },
+    {
+      "epoch": 1.941747572815534,
+      "grad_norm": 0.23946309089660645,
+      "learning_rate": 4.207687120701282e-05,
+      "loss": 0.0387,
+      "step": 1200
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9903582776377781,
+      "eval_f1": 0.8799067236688691,
+      "eval_loss": 0.04682581126689911,
+      "eval_precision": 0.8700999231360492,
+      "eval_recall": 0.889937106918239,
+      "eval_runtime": 2.8072,
+      "eval_samples_per_second": 439.943,
+      "eval_steps_per_second": 55.215,
+      "step": 1236
+    },
+    {
+      "epoch": 2.103559870550162,
+      "grad_norm": 0.8596442937850952,
+      "learning_rate": 4.0953023151269956e-05,
+      "loss": 0.0285,
+      "step": 1300
+    },
+    {
+      "epoch": 2.26537216828479,
+      "grad_norm": 0.03754520043730736,
+      "learning_rate": 3.9829175095527095e-05,
+      "loss": 0.0322,
+      "step": 1400
+    },
+    {
+      "epoch": 2.4271844660194173,
+      "grad_norm": 0.6684575080871582,
+      "learning_rate": 3.870532703978423e-05,
+      "loss": 0.023,
+      "step": 1500
+    },
+    {
+      "epoch": 2.588996763754045,
+      "grad_norm": 0.03833441436290741,
+      "learning_rate": 3.758147898404136e-05,
+      "loss": 0.0268,
+      "step": 1600
+    },
+    {
+      "epoch": 2.750809061488673,
+      "grad_norm": 0.3890291452407837,
+      "learning_rate": 3.6457630928298505e-05,
+      "loss": 0.0217,
+      "step": 1700
+    },
+    {
+      "epoch": 2.912621359223301,
+      "grad_norm": 0.4564450681209564,
+      "learning_rate": 3.533378287255564e-05,
+      "loss": 0.0295,
+      "step": 1800
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9906869727183083,
+      "eval_f1": 0.8855799373040752,
+      "eval_loss": 0.039505813270807266,
+      "eval_precision": 0.8828125,
+      "eval_recall": 0.8883647798742138,
+      "eval_runtime": 2.8133,
+      "eval_samples_per_second": 438.979,
+      "eval_steps_per_second": 55.095,
+      "step": 1854
+    },
+    {
+      "epoch": 3.074433656957929,
+      "grad_norm": 0.027059998363256454,
+      "learning_rate": 3.420993481681277e-05,
+      "loss": 0.0166,
+      "step": 1900
+    },
+    {
+      "epoch": 3.236245954692557,
+      "grad_norm": 0.030333412811160088,
+      "learning_rate": 3.308608676106991e-05,
+      "loss": 0.0174,
+      "step": 2000
+    },
+    {
+      "epoch": 3.3980582524271843,
+      "grad_norm": 0.13804250955581665,
+      "learning_rate": 3.196223870532705e-05,
+      "loss": 0.0153,
+      "step": 2100
+    },
+    {
+      "epoch": 3.559870550161812,
+      "grad_norm": 0.2849176824092865,
+      "learning_rate": 3.083839064958418e-05,
+      "loss": 0.0152,
+      "step": 2200
+    },
+    {
+      "epoch": 3.72168284789644,
+      "grad_norm": 0.14825651049613953,
+      "learning_rate": 2.971454259384132e-05,
+      "loss": 0.0171,
+      "step": 2300
+    },
+    {
+      "epoch": 3.883495145631068,
+      "grad_norm": 0.045380860567092896,
+      "learning_rate": 2.8590694538098453e-05,
+      "loss": 0.0255,
+      "step": 2400
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9920565355538512,
+      "eval_f1": 0.8999999999999999,
+      "eval_loss": 0.03599809855222702,
+      "eval_precision": 0.9014195583596214,
+      "eval_recall": 0.8985849056603774,
+      "eval_runtime": 2.8186,
+      "eval_samples_per_second": 438.161,
+      "eval_steps_per_second": 54.992,
+      "step": 2472
+    },
+    {
+      "epoch": 4.0453074433656955,
+      "grad_norm": 0.5658661723136902,
+      "learning_rate": 2.746684648235559e-05,
+      "loss": 0.0228,
+      "step": 2500
+    },
+    {
+      "epoch": 4.207119741100324,
+      "grad_norm": 0.11415175348520279,
+      "learning_rate": 2.6342998426612728e-05,
+      "loss": 0.0162,
+      "step": 2600
+    },
+    {
+      "epoch": 4.368932038834951,
+      "grad_norm": 0.1993759125471115,
+      "learning_rate": 2.5219150370869863e-05,
+      "loss": 0.0135,
+      "step": 2700
+    },
+    {
+      "epoch": 4.53074433656958,
+      "grad_norm": 0.11497118324041367,
+      "learning_rate": 2.4095302315127e-05,
+      "loss": 0.0159,
+      "step": 2800
+    },
+    {
+      "epoch": 4.692556634304207,
+      "grad_norm": 0.2147281914949417,
+      "learning_rate": 2.2971454259384134e-05,
+      "loss": 0.0156,
+      "step": 2900
+    },
+    {
+      "epoch": 4.854368932038835,
+      "grad_norm": 0.1083710715174675,
+      "learning_rate": 2.1847606203641273e-05,
+      "loss": 0.0094,
+      "step": 3000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9922756656075381,
+      "eval_f1": 0.9050980392156862,
+      "eval_loss": 0.03369523212313652,
+      "eval_precision": 0.9029733959311425,
+      "eval_recall": 0.9072327044025157,
+      "eval_runtime": 2.8037,
+      "eval_samples_per_second": 440.494,
+      "eval_steps_per_second": 55.285,
+      "step": 3090
+    },
+    {
+      "epoch": 5.016181229773463,
+      "grad_norm": 0.013677417300641537,
+      "learning_rate": 2.072375814789841e-05,
+      "loss": 0.016,
+      "step": 3100
+    },
+    {
+      "epoch": 5.17799352750809,
+      "grad_norm": 0.08207657188177109,
+      "learning_rate": 1.9599910092155544e-05,
+      "loss": 0.0133,
+      "step": 3200
+    },
+    {
+      "epoch": 5.339805825242719,
+      "grad_norm": 0.02103651873767376,
+      "learning_rate": 1.847606203641268e-05,
+      "loss": 0.0092,
+      "step": 3300
+    },
+    {
+      "epoch": 5.501618122977346,
+      "grad_norm": 1.4357458353042603,
+      "learning_rate": 1.735221398066982e-05,
+      "loss": 0.0122,
+      "step": 3400
+    },
+    {
+      "epoch": 5.663430420711974,
+      "grad_norm": 0.16999904811382294,
+      "learning_rate": 1.622836592492695e-05,
+      "loss": 0.0086,
+      "step": 3500
+    },
+    {
+      "epoch": 5.825242718446602,
+      "grad_norm": 0.09043747931718826,
+      "learning_rate": 1.510451786918409e-05,
+      "loss": 0.0093,
+      "step": 3600
+    },
+    {
+      "epoch": 5.9870550161812295,
+      "grad_norm": 0.06608462333679199,
+      "learning_rate": 1.3980669813441227e-05,
+      "loss": 0.0067,
+      "step": 3700
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9932617508491289,
+      "eval_f1": 0.9301960784313724,
+      "eval_loss": 0.033360060304403305,
+      "eval_precision": 0.9280125195618153,
+      "eval_recall": 0.9323899371069182,
+      "eval_runtime": 2.8189,
+      "eval_samples_per_second": 438.116,
+      "eval_steps_per_second": 54.986,
+      "step": 3708
+    },
+    {
+      "epoch": 6.148867313915858,
+      "grad_norm": 0.2284722775220871,
+      "learning_rate": 1.285682175769836e-05,
+      "loss": 0.0107,
+      "step": 3800
+    },
+    {
+      "epoch": 6.310679611650485,
+      "grad_norm": 0.02673812210559845,
+      "learning_rate": 1.1732973701955498e-05,
+      "loss": 0.0052,
+      "step": 3900
+    },
+    {
+      "epoch": 6.472491909385114,
+      "grad_norm": 0.33707210421562195,
+      "learning_rate": 1.0609125646212633e-05,
+      "loss": 0.0072,
+      "step": 4000
+    },
+    {
+      "epoch": 6.634304207119741,
+      "grad_norm": 0.0059865182265639305,
+      "learning_rate": 9.48527759046977e-06,
+      "loss": 0.0049,
+      "step": 4100
+    },
+    {
+      "epoch": 6.796116504854369,
+      "grad_norm": 0.2759881615638733,
+      "learning_rate": 8.361429534726907e-06,
+      "loss": 0.016,
+      "step": 4200
+    },
+    {
+      "epoch": 6.957928802588997,
+      "grad_norm": 0.18257270753383636,
+      "learning_rate": 7.237581478984042e-06,
+      "loss": 0.0069,
+      "step": 4300
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.9937000109565027,
+      "eval_f1": 0.935483870967742,
+      "eval_loss": 0.0347304567694664,
+      "eval_precision": 0.9362204724409449,
+      "eval_recall": 0.934748427672956,
+      "eval_runtime": 2.8106,
+      "eval_samples_per_second": 439.402,
+      "eval_steps_per_second": 55.148,
+      "step": 4326
+    },
+    {
+      "epoch": 7.119741100323624,
+      "grad_norm": 0.007623529061675072,
+      "learning_rate": 6.113733423241179e-06,
+      "loss": 0.0046,
+      "step": 4400
+    },
+    {
+      "epoch": 7.281553398058253,
+      "grad_norm": 0.043167050927877426,
+      "learning_rate": 4.989885367498316e-06,
+      "loss": 0.009,
+      "step": 4500
+    },
+    {
+      "epoch": 7.44336569579288,
+      "grad_norm": 0.009674232453107834,
+      "learning_rate": 3.866037311755451e-06,
+      "loss": 0.0046,
+      "step": 4600
+    },
+    {
+      "epoch": 7.605177993527509,
+      "grad_norm": 0.05575043708086014,
+      "learning_rate": 2.742189256012588e-06,
+      "loss": 0.0052,
+      "step": 4700
+    },
+    {
+      "epoch": 7.766990291262136,
+      "grad_norm": 0.006715767551213503,
+      "learning_rate": 1.6183412002697239e-06,
+      "loss": 0.0044,
+      "step": 4800
+    },
+    {
+      "epoch": 7.9288025889967635,
+      "grad_norm": 0.009280543774366379,
+      "learning_rate": 4.9449314452686e-07,
+      "loss": 0.0054,
+      "step": 4900
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.9936726196997918,
+      "eval_f1": 0.93401413982718,
+      "eval_loss": 0.03279910609126091,
+      "eval_precision": 0.9332810047095761,
+      "eval_recall": 0.934748427672956,
+      "eval_runtime": 2.829,
+      "eval_samples_per_second": 436.551,
+      "eval_steps_per_second": 54.79,
+      "step": 4944
+    },
+    {
+      "epoch": 8.0,
+      "step": 4944,
+      "total_flos": 2833132740217920.0,
+      "train_loss": 0.08807948804957774,
+      "train_runtime": 679.3683,
+      "train_samples_per_second": 116.355,
+      "train_steps_per_second": 7.277
+    }
+  ],
+  "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": 2833132740217920.0,
+  "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 sha256:5a7a9b89b841739a3021e1e3a1adf2c68602fa7afc6688bbb60862c85d3c4c5e
+size 5624