Upload 11 files
Browse files- config.json +35 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +58 -0
- trainer_state.json +337 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "bioformers/bioformer-16L",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForTokenClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 384,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "I-mutant",
|
| 13 |
+
"1": "B-mutant",
|
| 14 |
+
"2": "O"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 1536,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"B-mutant": 1,
|
| 20 |
+
"I-mutant": 0,
|
| 21 |
+
"O": 2
|
| 22 |
+
},
|
| 23 |
+
"layer_norm_eps": 1e-12,
|
| 24 |
+
"max_position_embeddings": 1024,
|
| 25 |
+
"model_type": "bert",
|
| 26 |
+
"num_attention_heads": 6,
|
| 27 |
+
"num_hidden_layers": 16,
|
| 28 |
+
"pad_token_id": 0,
|
| 29 |
+
"position_embedding_type": "absolute",
|
| 30 |
+
"torch_dtype": "float32",
|
| 31 |
+
"transformers_version": "4.48.2",
|
| 32 |
+
"type_vocab_size": 2,
|
| 33 |
+
"use_cache": true,
|
| 34 |
+
"vocab_size": 32768
|
| 35 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e3dd6de04a00b51d8602340de1e4b2f7048855f92caac805102b70c4d663ea1b
|
| 3 |
+
size 165511700
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce95340dfb5860e4ac6ac43bd3e001eeb687fe5da0c1c0e312f4d91613a841e1
|
| 3 |
+
size 331181818
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7831b301db3133cd9e359e7a5822571431ce40de8c220a797efed0ff1f54d777
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9edcc85a98c6e6388be55dec98392c30840bdbe7b9fbfdaf6ff7fa30afa38256
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": false,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 51 |
+
"never_split": null,
|
| 52 |
+
"pad_token": "[PAD]",
|
| 53 |
+
"sep_token": "[SEP]",
|
| 54 |
+
"strip_accents": null,
|
| 55 |
+
"tokenize_chinese_chars": true,
|
| 56 |
+
"tokenizer_class": "BertTokenizer",
|
| 57 |
+
"unk_token": "[UNK]"
|
| 58 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,337 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8884462151394423,
|
| 3 |
+
"best_model_checkpoint": "/hps/software/users/chembl/christine/git_projects/OTAR3088/Entity-Extraction-Modular-pipeline/outputs/model_outputs/hf/Variants_V1/base/bioformers/checkpoint-464",
|
| 4 |
+
"epoch": 16.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 464,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"grad_norm": 2.3756778240203857,
|
| 14 |
+
"learning_rate": 9.310344827586207e-06,
|
| 15 |
+
"loss": 0.7473,
|
| 16 |
+
"step": 29
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_accuracy": 0.8869123252858958,
|
| 21 |
+
"eval_f1": 0.0,
|
| 22 |
+
"eval_loss": 0.4901806712150574,
|
| 23 |
+
"eval_precision": 0.0,
|
| 24 |
+
"eval_recall": 0.0,
|
| 25 |
+
"eval_runtime": 1.1486,
|
| 26 |
+
"eval_samples_per_second": 87.93,
|
| 27 |
+
"eval_steps_per_second": 6.094,
|
| 28 |
+
"step": 29
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 2.0,
|
| 32 |
+
"grad_norm": 0.7127112746238708,
|
| 33 |
+
"learning_rate": 1.931034482758621e-05,
|
| 34 |
+
"loss": 0.3253,
|
| 35 |
+
"step": 58
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 2.0,
|
| 39 |
+
"eval_accuracy": 0.9199491740787802,
|
| 40 |
+
"eval_f1": 0.2011173184357542,
|
| 41 |
+
"eval_loss": 0.2195604145526886,
|
| 42 |
+
"eval_precision": 0.3050847457627119,
|
| 43 |
+
"eval_recall": 0.15,
|
| 44 |
+
"eval_runtime": 0.9154,
|
| 45 |
+
"eval_samples_per_second": 110.332,
|
| 46 |
+
"eval_steps_per_second": 7.647,
|
| 47 |
+
"step": 58
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 3.0,
|
| 51 |
+
"grad_norm": 1.941286563873291,
|
| 52 |
+
"learning_rate": 1.896551724137931e-05,
|
| 53 |
+
"loss": 0.129,
|
| 54 |
+
"step": 87
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.0,
|
| 58 |
+
"eval_accuracy": 0.9618805590851334,
|
| 59 |
+
"eval_f1": 0.741747572815534,
|
| 60 |
+
"eval_loss": 0.10903950780630112,
|
| 61 |
+
"eval_precision": 0.6945454545454546,
|
| 62 |
+
"eval_recall": 0.7958333333333333,
|
| 63 |
+
"eval_runtime": 0.9369,
|
| 64 |
+
"eval_samples_per_second": 107.801,
|
| 65 |
+
"eval_steps_per_second": 7.471,
|
| 66 |
+
"step": 87
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 4.0,
|
| 70 |
+
"grad_norm": 1.2016586065292358,
|
| 71 |
+
"learning_rate": 1.78544061302682e-05,
|
| 72 |
+
"loss": 0.0581,
|
| 73 |
+
"step": 116
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 4.0,
|
| 77 |
+
"eval_accuracy": 0.974587039390089,
|
| 78 |
+
"eval_f1": 0.8183556405353729,
|
| 79 |
+
"eval_loss": 0.08683695644140244,
|
| 80 |
+
"eval_precision": 0.7561837455830389,
|
| 81 |
+
"eval_recall": 0.8916666666666667,
|
| 82 |
+
"eval_runtime": 0.9184,
|
| 83 |
+
"eval_samples_per_second": 109.975,
|
| 84 |
+
"eval_steps_per_second": 7.622,
|
| 85 |
+
"step": 116
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 5.0,
|
| 89 |
+
"grad_norm": 1.6464694738388062,
|
| 90 |
+
"learning_rate": 1.674329501915709e-05,
|
| 91 |
+
"loss": 0.0365,
|
| 92 |
+
"step": 145
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 5.0,
|
| 96 |
+
"eval_accuracy": 0.9760694620923338,
|
| 97 |
+
"eval_f1": 0.8285163776493256,
|
| 98 |
+
"eval_loss": 0.07583592087030411,
|
| 99 |
+
"eval_precision": 0.7706093189964157,
|
| 100 |
+
"eval_recall": 0.8958333333333334,
|
| 101 |
+
"eval_runtime": 0.9273,
|
| 102 |
+
"eval_samples_per_second": 108.916,
|
| 103 |
+
"eval_steps_per_second": 7.549,
|
| 104 |
+
"step": 145
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.0,
|
| 108 |
+
"grad_norm": 0.19638490676879883,
|
| 109 |
+
"learning_rate": 1.563218390804598e-05,
|
| 110 |
+
"loss": 0.0251,
|
| 111 |
+
"step": 174
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 6.0,
|
| 115 |
+
"eval_accuracy": 0.9788225328250741,
|
| 116 |
+
"eval_f1": 0.8470588235294118,
|
| 117 |
+
"eval_loss": 0.06711125373840332,
|
| 118 |
+
"eval_precision": 0.8,
|
| 119 |
+
"eval_recall": 0.9,
|
| 120 |
+
"eval_runtime": 0.9219,
|
| 121 |
+
"eval_samples_per_second": 109.558,
|
| 122 |
+
"eval_steps_per_second": 7.593,
|
| 123 |
+
"step": 174
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 7.0,
|
| 127 |
+
"grad_norm": 0.5702025294303894,
|
| 128 |
+
"learning_rate": 1.4521072796934867e-05,
|
| 129 |
+
"loss": 0.0153,
|
| 130 |
+
"step": 203
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 7.0,
|
| 134 |
+
"eval_accuracy": 0.9800931808555696,
|
| 135 |
+
"eval_f1": 0.8649706457925636,
|
| 136 |
+
"eval_loss": 0.06598453223705292,
|
| 137 |
+
"eval_precision": 0.8154981549815498,
|
| 138 |
+
"eval_recall": 0.9208333333333333,
|
| 139 |
+
"eval_runtime": 0.9157,
|
| 140 |
+
"eval_samples_per_second": 110.293,
|
| 141 |
+
"eval_steps_per_second": 7.644,
|
| 142 |
+
"step": 203
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 8.0,
|
| 146 |
+
"grad_norm": 0.4013058543205261,
|
| 147 |
+
"learning_rate": 1.3409961685823755e-05,
|
| 148 |
+
"loss": 0.0107,
|
| 149 |
+
"step": 232
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 8.0,
|
| 153 |
+
"eval_accuracy": 0.9815756035578145,
|
| 154 |
+
"eval_f1": 0.8678500986193293,
|
| 155 |
+
"eval_loss": 0.06503612548112869,
|
| 156 |
+
"eval_precision": 0.8239700374531835,
|
| 157 |
+
"eval_recall": 0.9166666666666666,
|
| 158 |
+
"eval_runtime": 0.9192,
|
| 159 |
+
"eval_samples_per_second": 109.876,
|
| 160 |
+
"eval_steps_per_second": 7.615,
|
| 161 |
+
"step": 232
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 9.0,
|
| 165 |
+
"grad_norm": 0.62887042760849,
|
| 166 |
+
"learning_rate": 1.2298850574712644e-05,
|
| 167 |
+
"loss": 0.0096,
|
| 168 |
+
"step": 261
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 9.0,
|
| 172 |
+
"eval_accuracy": 0.9781872088098263,
|
| 173 |
+
"eval_f1": 0.8727984344422701,
|
| 174 |
+
"eval_loss": 0.09074747562408447,
|
| 175 |
+
"eval_precision": 0.8228782287822878,
|
| 176 |
+
"eval_recall": 0.9291666666666667,
|
| 177 |
+
"eval_runtime": 0.9179,
|
| 178 |
+
"eval_samples_per_second": 110.04,
|
| 179 |
+
"eval_steps_per_second": 7.627,
|
| 180 |
+
"step": 261
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 10.0,
|
| 184 |
+
"grad_norm": 0.13981495797634125,
|
| 185 |
+
"learning_rate": 1.1187739463601532e-05,
|
| 186 |
+
"loss": 0.0075,
|
| 187 |
+
"step": 290
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 10.0,
|
| 191 |
+
"eval_accuracy": 0.9800931808555696,
|
| 192 |
+
"eval_f1": 0.8638132295719844,
|
| 193 |
+
"eval_loss": 0.07469259202480316,
|
| 194 |
+
"eval_precision": 0.8102189781021898,
|
| 195 |
+
"eval_recall": 0.925,
|
| 196 |
+
"eval_runtime": 0.8265,
|
| 197 |
+
"eval_samples_per_second": 122.199,
|
| 198 |
+
"eval_steps_per_second": 8.469,
|
| 199 |
+
"step": 290
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 11.0,
|
| 203 |
+
"grad_norm": 0.29558104276657104,
|
| 204 |
+
"learning_rate": 1.0076628352490422e-05,
|
| 205 |
+
"loss": 0.007,
|
| 206 |
+
"step": 319
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 11.0,
|
| 210 |
+
"eval_accuracy": 0.9794578568403219,
|
| 211 |
+
"eval_f1": 0.8621359223300971,
|
| 212 |
+
"eval_loss": 0.07568900287151337,
|
| 213 |
+
"eval_precision": 0.8072727272727273,
|
| 214 |
+
"eval_recall": 0.925,
|
| 215 |
+
"eval_runtime": 0.8257,
|
| 216 |
+
"eval_samples_per_second": 122.315,
|
| 217 |
+
"eval_steps_per_second": 8.477,
|
| 218 |
+
"step": 319
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 12.0,
|
| 222 |
+
"grad_norm": 0.15115247666835785,
|
| 223 |
+
"learning_rate": 8.965517241379312e-06,
|
| 224 |
+
"loss": 0.0059,
|
| 225 |
+
"step": 348
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 12.0,
|
| 229 |
+
"eval_accuracy": 0.9828462515883101,
|
| 230 |
+
"eval_f1": 0.8695652173913043,
|
| 231 |
+
"eval_loss": 0.07132618129253387,
|
| 232 |
+
"eval_precision": 0.8270676691729323,
|
| 233 |
+
"eval_recall": 0.9166666666666666,
|
| 234 |
+
"eval_runtime": 0.832,
|
| 235 |
+
"eval_samples_per_second": 121.396,
|
| 236 |
+
"eval_steps_per_second": 8.414,
|
| 237 |
+
"step": 348
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 13.0,
|
| 241 |
+
"grad_norm": 0.05523902550339699,
|
| 242 |
+
"learning_rate": 7.854406130268199e-06,
|
| 243 |
+
"loss": 0.0055,
|
| 244 |
+
"step": 377
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 13.0,
|
| 248 |
+
"eval_accuracy": 0.9822109275730623,
|
| 249 |
+
"eval_f1": 0.8848484848484848,
|
| 250 |
+
"eval_loss": 0.07388192415237427,
|
| 251 |
+
"eval_precision": 0.8588235294117647,
|
| 252 |
+
"eval_recall": 0.9125,
|
| 253 |
+
"eval_runtime": 0.8308,
|
| 254 |
+
"eval_samples_per_second": 121.568,
|
| 255 |
+
"eval_steps_per_second": 8.425,
|
| 256 |
+
"step": 377
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 14.0,
|
| 260 |
+
"grad_norm": 0.08825332671403885,
|
| 261 |
+
"learning_rate": 6.743295019157089e-06,
|
| 262 |
+
"loss": 0.0048,
|
| 263 |
+
"step": 406
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 14.0,
|
| 267 |
+
"eval_accuracy": 0.9822109275730623,
|
| 268 |
+
"eval_f1": 0.8717948717948718,
|
| 269 |
+
"eval_loss": 0.07594037801027298,
|
| 270 |
+
"eval_precision": 0.8277153558052435,
|
| 271 |
+
"eval_recall": 0.9208333333333333,
|
| 272 |
+
"eval_runtime": 0.9402,
|
| 273 |
+
"eval_samples_per_second": 107.427,
|
| 274 |
+
"eval_steps_per_second": 7.445,
|
| 275 |
+
"step": 406
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 15.0,
|
| 279 |
+
"grad_norm": 0.029169419780373573,
|
| 280 |
+
"learning_rate": 5.6321839080459775e-06,
|
| 281 |
+
"loss": 0.0051,
|
| 282 |
+
"step": 435
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 15.0,
|
| 286 |
+
"eval_accuracy": 0.981152054214316,
|
| 287 |
+
"eval_f1": 0.8858267716535433,
|
| 288 |
+
"eval_loss": 0.08297453075647354,
|
| 289 |
+
"eval_precision": 0.8395522388059702,
|
| 290 |
+
"eval_recall": 0.9375,
|
| 291 |
+
"eval_runtime": 0.9214,
|
| 292 |
+
"eval_samples_per_second": 109.612,
|
| 293 |
+
"eval_steps_per_second": 7.597,
|
| 294 |
+
"step": 435
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 16.0,
|
| 298 |
+
"grad_norm": 0.1920151710510254,
|
| 299 |
+
"learning_rate": 4.521072796934866e-06,
|
| 300 |
+
"loss": 0.0039,
|
| 301 |
+
"step": 464
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 16.0,
|
| 305 |
+
"eval_accuracy": 0.9832698009318086,
|
| 306 |
+
"eval_f1": 0.8884462151394423,
|
| 307 |
+
"eval_loss": 0.07141096144914627,
|
| 308 |
+
"eval_precision": 0.851145038167939,
|
| 309 |
+
"eval_recall": 0.9291666666666667,
|
| 310 |
+
"eval_runtime": 0.9091,
|
| 311 |
+
"eval_samples_per_second": 111.098,
|
| 312 |
+
"eval_steps_per_second": 7.7,
|
| 313 |
+
"step": 464
|
| 314 |
+
}
|
| 315 |
+
],
|
| 316 |
+
"logging_steps": 500,
|
| 317 |
+
"max_steps": 580,
|
| 318 |
+
"num_input_tokens_seen": 0,
|
| 319 |
+
"num_train_epochs": 20,
|
| 320 |
+
"save_steps": 500,
|
| 321 |
+
"stateful_callbacks": {
|
| 322 |
+
"TrainerControl": {
|
| 323 |
+
"args": {
|
| 324 |
+
"should_epoch_stop": false,
|
| 325 |
+
"should_evaluate": false,
|
| 326 |
+
"should_log": false,
|
| 327 |
+
"should_save": true,
|
| 328 |
+
"should_training_stop": false
|
| 329 |
+
},
|
| 330 |
+
"attributes": {}
|
| 331 |
+
}
|
| 332 |
+
},
|
| 333 |
+
"total_flos": 286480239652512.0,
|
| 334 |
+
"train_batch_size": 16,
|
| 335 |
+
"trial_name": null,
|
| 336 |
+
"trial_params": null
|
| 337 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f44f8d23d64c562c40644bb0b0cd577ea47961d1c2957fa9b9dc703491b3b29c
|
| 3 |
+
size 5560
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|