cite-text-analysis
commited on
Commit
•
456b246
1
Parent(s):
0139801
Upload folder using huggingface_hub
Browse files- training_checkpoints/checkpoint-4032/added_tokens.json +3 -0
- training_checkpoints/checkpoint-4032/config.json +41 -0
- training_checkpoints/checkpoint-4032/model.safetensors +3 -0
- training_checkpoints/checkpoint-4032/optimizer.pt +3 -0
- training_checkpoints/checkpoint-4032/rng_state.pth +3 -0
- training_checkpoints/checkpoint-4032/scheduler.pt +3 -0
- training_checkpoints/checkpoint-4032/special_tokens_map.json +13 -0
- training_checkpoints/checkpoint-4032/tokenizer.json +0 -0
- training_checkpoints/checkpoint-4032/tokenizer_config.json +65 -0
- training_checkpoints/checkpoint-4032/trainer_state.json +473 -0
- training_checkpoints/checkpoint-4032/training_args.bin +3 -0
- training_checkpoints/checkpoint-4032/vocab.txt +0 -0
training_checkpoints/checkpoint-4032/added_tokens.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<pad>": 30522
|
3 |
+
}
|
training_checkpoints/checkpoint-4032/config.json
ADDED
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "nlpaueb/legal-bert-base-uncased",
|
3 |
+
"architectures": [
|
4 |
+
"BertForSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"bos_token_id": 0,
|
8 |
+
"classifier_dropout": null,
|
9 |
+
"eos_token_ids": 0,
|
10 |
+
"hidden_act": "gelu",
|
11 |
+
"hidden_dropout_prob": 0.1,
|
12 |
+
"hidden_size": 768,
|
13 |
+
"id2label": {
|
14 |
+
"0": "positive",
|
15 |
+
"1": "negative",
|
16 |
+
"2": "neutral",
|
17 |
+
"3": "rel_by_party"
|
18 |
+
},
|
19 |
+
"initializer_range": 0.02,
|
20 |
+
"intermediate_size": 3072,
|
21 |
+
"label2id": {
|
22 |
+
"negative": 1,
|
23 |
+
"neutral": 2,
|
24 |
+
"positive": 0,
|
25 |
+
"rel_by_party": 3
|
26 |
+
},
|
27 |
+
"layer_norm_eps": 1e-12,
|
28 |
+
"max_position_embeddings": 512,
|
29 |
+
"model_type": "bert",
|
30 |
+
"num_attention_heads": 12,
|
31 |
+
"num_hidden_layers": 12,
|
32 |
+
"output_past": true,
|
33 |
+
"pad_token_id": 30522,
|
34 |
+
"position_embedding_type": "absolute",
|
35 |
+
"problem_type": "single_label_classification",
|
36 |
+
"torch_dtype": "float32",
|
37 |
+
"transformers_version": "4.40.2",
|
38 |
+
"type_vocab_size": 2,
|
39 |
+
"use_cache": true,
|
40 |
+
"vocab_size": 30523
|
41 |
+
}
|
training_checkpoints/checkpoint-4032/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:34333fb555be53e0448eb786aadc36b479addf407ffc5038cb208d00bb83682d
|
3 |
+
size 437967872
|
training_checkpoints/checkpoint-4032/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4b7ba7851daeb4fc0e5337a3123cc762f1aa914d9f66858438bb2f93aec7ff71
|
3 |
+
size 876056826
|
training_checkpoints/checkpoint-4032/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:90b9864635816b65a9d0800901169060e3eb05c2210682bb7259d374e5d72ce7
|
3 |
+
size 14244
|
training_checkpoints/checkpoint-4032/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dee6bfb1792da04ffab8dd58e17618838cb8ea5c97277055e98cf1431b58baa7
|
3 |
+
size 1064
|
training_checkpoints/checkpoint-4032/special_tokens_map.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "[CLS]",
|
3 |
+
"mask_token": "[MASK]",
|
4 |
+
"pad_token": {
|
5 |
+
"content": "<pad>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": false,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false
|
10 |
+
},
|
11 |
+
"sep_token": "[SEP]",
|
12 |
+
"unk_token": "[UNK]"
|
13 |
+
}
|
training_checkpoints/checkpoint-4032/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
training_checkpoints/checkpoint-4032/tokenizer_config.json
ADDED
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"30522": {
|
44 |
+
"content": "<pad>",
|
45 |
+
"lstrip": false,
|
46 |
+
"normalized": false,
|
47 |
+
"rstrip": false,
|
48 |
+
"single_word": false,
|
49 |
+
"special": true
|
50 |
+
}
|
51 |
+
},
|
52 |
+
"clean_up_tokenization_spaces": true,
|
53 |
+
"cls_token": "[CLS]",
|
54 |
+
"do_basic_tokenize": true,
|
55 |
+
"do_lower_case": true,
|
56 |
+
"mask_token": "[MASK]",
|
57 |
+
"model_max_length": 512,
|
58 |
+
"never_split": null,
|
59 |
+
"pad_token": "<pad>",
|
60 |
+
"sep_token": "[SEP]",
|
61 |
+
"strip_accents": null,
|
62 |
+
"tokenize_chinese_chars": true,
|
63 |
+
"tokenizer_class": "BertTokenizer",
|
64 |
+
"unk_token": "[UNK]"
|
65 |
+
}
|
training_checkpoints/checkpoint-4032/trainer_state.json
ADDED
@@ -0,0 +1,473 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.6764229637898442,
|
3 |
+
"best_model_checkpoint": "case-analysis-legal-bert-base-uncased/checkpoint-2464",
|
4 |
+
"epoch": 18.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 4032,
|
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 |
+
"eval_accuracy": 0.6570155902004454,
|
14 |
+
"eval_f1_macro": 0.4081263564022185,
|
15 |
+
"eval_f1_micro": 0.6570155902004454,
|
16 |
+
"eval_f1_weighted": 0.6166399839312825,
|
17 |
+
"eval_loss": 1.0114229917526245,
|
18 |
+
"eval_macro_fpr": 0.19475132363363487,
|
19 |
+
"eval_macro_sensitivity": 0.40746667217255456,
|
20 |
+
"eval_macro_specificity": 0.8324392236563901,
|
21 |
+
"eval_precision": 0.6178816225558943,
|
22 |
+
"eval_precision_macro": 0.4664000256488353,
|
23 |
+
"eval_recall": 0.6570155902004454,
|
24 |
+
"eval_recall_macro": 0.40746667217255456,
|
25 |
+
"eval_runtime": 7.3199,
|
26 |
+
"eval_samples_per_second": 61.34,
|
27 |
+
"eval_steps_per_second": 7.787,
|
28 |
+
"eval_weighted_fpr": 0.1482194417709336,
|
29 |
+
"eval_weighted_sensitivity": 0.6570155902004454,
|
30 |
+
"eval_weighted_specificity": 0.6727413044251147,
|
31 |
+
"step": 224
|
32 |
+
},
|
33 |
+
{
|
34 |
+
"epoch": 2.0,
|
35 |
+
"eval_accuracy": 0.7750556792873051,
|
36 |
+
"eval_f1_macro": 0.5658663855845003,
|
37 |
+
"eval_f1_micro": 0.775055679287305,
|
38 |
+
"eval_f1_weighted": 0.7564492543508543,
|
39 |
+
"eval_loss": 0.764952540397644,
|
40 |
+
"eval_macro_fpr": 0.10943701020063132,
|
41 |
+
"eval_macro_sensitivity": 0.5805958747135218,
|
42 |
+
"eval_macro_specificity": 0.9039084791647444,
|
43 |
+
"eval_precision": 0.7425093237919278,
|
44 |
+
"eval_precision_macro": 0.5566158325717437,
|
45 |
+
"eval_recall": 0.7750556792873051,
|
46 |
+
"eval_recall_macro": 0.5805958747135218,
|
47 |
+
"eval_runtime": 30.7218,
|
48 |
+
"eval_samples_per_second": 14.615,
|
49 |
+
"eval_steps_per_second": 1.855,
|
50 |
+
"eval_weighted_fpr": 0.08820960698689956,
|
51 |
+
"eval_weighted_sensitivity": 0.7750556792873051,
|
52 |
+
"eval_weighted_specificity": 0.8405782373716723,
|
53 |
+
"step": 448
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 2.232142857142857,
|
57 |
+
"grad_norm": 9.53757095336914,
|
58 |
+
"learning_rate": 4.631696428571429e-05,
|
59 |
+
"loss": 0.7677,
|
60 |
+
"step": 500
|
61 |
+
},
|
62 |
+
{
|
63 |
+
"epoch": 3.0,
|
64 |
+
"eval_accuracy": 0.7817371937639198,
|
65 |
+
"eval_f1_macro": 0.5707248076329534,
|
66 |
+
"eval_f1_micro": 0.7817371937639198,
|
67 |
+
"eval_f1_weighted": 0.7678448816993292,
|
68 |
+
"eval_loss": 0.7341566681861877,
|
69 |
+
"eval_macro_fpr": 0.10409605667370833,
|
70 |
+
"eval_macro_sensitivity": 0.5967117805353099,
|
71 |
+
"eval_macro_specificity": 0.9083046086144851,
|
72 |
+
"eval_precision": 0.7694906286035585,
|
73 |
+
"eval_precision_macro": 0.5673619923602318,
|
74 |
+
"eval_recall": 0.7817371937639198,
|
75 |
+
"eval_recall_macro": 0.5967117805353099,
|
76 |
+
"eval_runtime": 32.9882,
|
77 |
+
"eval_samples_per_second": 13.611,
|
78 |
+
"eval_steps_per_second": 1.728,
|
79 |
+
"eval_weighted_fpr": 0.08514335360556038,
|
80 |
+
"eval_weighted_sensitivity": 0.7817371937639198,
|
81 |
+
"eval_weighted_specificity": 0.8514812406940205,
|
82 |
+
"step": 672
|
83 |
+
},
|
84 |
+
{
|
85 |
+
"epoch": 4.0,
|
86 |
+
"eval_accuracy": 0.8173719376391982,
|
87 |
+
"eval_f1_macro": 0.5903876477694423,
|
88 |
+
"eval_f1_micro": 0.8173719376391981,
|
89 |
+
"eval_f1_weighted": 0.7920921547144288,
|
90 |
+
"eval_loss": 0.7967568635940552,
|
91 |
+
"eval_macro_fpr": 0.09654289346260755,
|
92 |
+
"eval_macro_sensitivity": 0.5892796922208687,
|
93 |
+
"eval_macro_specificity": 0.9135512326349768,
|
94 |
+
"eval_precision": 0.776630887817645,
|
95 |
+
"eval_precision_macro": 0.6035747039094435,
|
96 |
+
"eval_recall": 0.8173719376391982,
|
97 |
+
"eval_recall_macro": 0.5892796922208687,
|
98 |
+
"eval_runtime": 31.5737,
|
99 |
+
"eval_samples_per_second": 14.221,
|
100 |
+
"eval_steps_per_second": 1.805,
|
101 |
+
"eval_weighted_fpr": 0.06931530008453085,
|
102 |
+
"eval_weighted_sensitivity": 0.8173719376391982,
|
103 |
+
"eval_weighted_specificity": 0.836832992900709,
|
104 |
+
"step": 896
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 4.464285714285714,
|
108 |
+
"grad_norm": 0.4865407943725586,
|
109 |
+
"learning_rate": 4.260416666666667e-05,
|
110 |
+
"loss": 0.482,
|
111 |
+
"step": 1000
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 5.0,
|
115 |
+
"eval_accuracy": 0.8084632516703786,
|
116 |
+
"eval_f1_macro": 0.6140549414742963,
|
117 |
+
"eval_f1_micro": 0.8084632516703786,
|
118 |
+
"eval_f1_weighted": 0.7940411816552344,
|
119 |
+
"eval_loss": 0.8170605301856995,
|
120 |
+
"eval_macro_fpr": 0.09903328320796416,
|
121 |
+
"eval_macro_sensitivity": 0.6038384641325818,
|
122 |
+
"eval_macro_specificity": 0.9107621964110777,
|
123 |
+
"eval_precision": 0.7852994506060567,
|
124 |
+
"eval_precision_macro": 0.6365555789178582,
|
125 |
+
"eval_recall": 0.8084632516703786,
|
126 |
+
"eval_recall_macro": 0.6038384641325818,
|
127 |
+
"eval_runtime": 33.691,
|
128 |
+
"eval_samples_per_second": 13.327,
|
129 |
+
"eval_steps_per_second": 1.692,
|
130 |
+
"eval_weighted_fpr": 0.07319148936170213,
|
131 |
+
"eval_weighted_sensitivity": 0.8084632516703786,
|
132 |
+
"eval_weighted_specificity": 0.8345855339739323,
|
133 |
+
"step": 1120
|
134 |
+
},
|
135 |
+
{
|
136 |
+
"epoch": 6.0,
|
137 |
+
"eval_accuracy": 0.8240534521158129,
|
138 |
+
"eval_f1_macro": 0.6314846316884821,
|
139 |
+
"eval_f1_micro": 0.8240534521158129,
|
140 |
+
"eval_f1_weighted": 0.8083544635272152,
|
141 |
+
"eval_loss": 0.8909738659858704,
|
142 |
+
"eval_macro_fpr": 0.087529695086195,
|
143 |
+
"eval_macro_sensitivity": 0.6358500402618049,
|
144 |
+
"eval_macro_specificity": 0.9211652121242622,
|
145 |
+
"eval_precision": 0.8028335761470152,
|
146 |
+
"eval_precision_macro": 0.6659870176540079,
|
147 |
+
"eval_recall": 0.8240534521158129,
|
148 |
+
"eval_recall_macro": 0.6358500402618049,
|
149 |
+
"eval_runtime": 32.5936,
|
150 |
+
"eval_samples_per_second": 13.776,
|
151 |
+
"eval_steps_per_second": 1.749,
|
152 |
+
"eval_weighted_fpr": 0.06644238856181665,
|
153 |
+
"eval_weighted_sensitivity": 0.8240534521158129,
|
154 |
+
"eval_weighted_specificity": 0.8606073963812356,
|
155 |
+
"step": 1344
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 6.696428571428571,
|
159 |
+
"grad_norm": 24.31475257873535,
|
160 |
+
"learning_rate": 3.888392857142857e-05,
|
161 |
+
"loss": 0.2993,
|
162 |
+
"step": 1500
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 7.0,
|
166 |
+
"eval_accuracy": 0.8040089086859689,
|
167 |
+
"eval_f1_macro": 0.6741993002516824,
|
168 |
+
"eval_f1_micro": 0.8040089086859689,
|
169 |
+
"eval_f1_weighted": 0.8067155865282774,
|
170 |
+
"eval_loss": 1.0094465017318726,
|
171 |
+
"eval_macro_fpr": 0.08913550870826559,
|
172 |
+
"eval_macro_sensitivity": 0.6773721773721774,
|
173 |
+
"eval_macro_specificity": 0.9217242184731483,
|
174 |
+
"eval_precision": 0.8108961670502116,
|
175 |
+
"eval_precision_macro": 0.6741723864688478,
|
176 |
+
"eval_recall": 0.8040089086859689,
|
177 |
+
"eval_recall_macro": 0.6773721773721774,
|
178 |
+
"eval_runtime": 33.2368,
|
179 |
+
"eval_samples_per_second": 13.509,
|
180 |
+
"eval_steps_per_second": 1.715,
|
181 |
+
"eval_weighted_fpr": 0.07514944491887275,
|
182 |
+
"eval_weighted_sensitivity": 0.8040089086859689,
|
183 |
+
"eval_weighted_specificity": 0.8828879652066246,
|
184 |
+
"step": 1568
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 8.0,
|
188 |
+
"eval_accuracy": 0.8106904231625836,
|
189 |
+
"eval_f1_macro": 0.6229381248976396,
|
190 |
+
"eval_f1_micro": 0.8106904231625834,
|
191 |
+
"eval_f1_weighted": 0.8016368957819484,
|
192 |
+
"eval_loss": 1.1504098176956177,
|
193 |
+
"eval_macro_fpr": 0.089652400577408,
|
194 |
+
"eval_macro_sensitivity": 0.6330299933241109,
|
195 |
+
"eval_macro_specificity": 0.9203722429802373,
|
196 |
+
"eval_precision": 0.7967862579587831,
|
197 |
+
"eval_precision_macro": 0.6227581659565835,
|
198 |
+
"eval_recall": 0.8106904231625836,
|
199 |
+
"eval_recall_macro": 0.6330299933241109,
|
200 |
+
"eval_runtime": 36.2257,
|
201 |
+
"eval_samples_per_second": 12.395,
|
202 |
+
"eval_steps_per_second": 1.573,
|
203 |
+
"eval_weighted_fpr": 0.07221750212404418,
|
204 |
+
"eval_weighted_sensitivity": 0.8106904231625836,
|
205 |
+
"eval_weighted_specificity": 0.8707985487583659,
|
206 |
+
"step": 1792
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 8.928571428571429,
|
210 |
+
"grad_norm": 0.21054323017597198,
|
211 |
+
"learning_rate": 3.516369047619048e-05,
|
212 |
+
"loss": 0.1367,
|
213 |
+
"step": 2000
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 9.0,
|
217 |
+
"eval_accuracy": 0.8062360801781737,
|
218 |
+
"eval_f1_macro": 0.6359771704966781,
|
219 |
+
"eval_f1_micro": 0.8062360801781738,
|
220 |
+
"eval_f1_weighted": 0.8044711013673843,
|
221 |
+
"eval_loss": 1.2533091306686401,
|
222 |
+
"eval_macro_fpr": 0.09482503678094849,
|
223 |
+
"eval_macro_sensitivity": 0.6224819165995636,
|
224 |
+
"eval_macro_specificity": 0.9164236762016751,
|
225 |
+
"eval_precision": 0.8059392062620114,
|
226 |
+
"eval_precision_macro": 0.6536979385970869,
|
227 |
+
"eval_recall": 0.8062360801781737,
|
228 |
+
"eval_recall_macro": 0.6224819165995636,
|
229 |
+
"eval_runtime": 30.7582,
|
230 |
+
"eval_samples_per_second": 14.598,
|
231 |
+
"eval_steps_per_second": 1.853,
|
232 |
+
"eval_weighted_fpr": 0.0741687979539642,
|
233 |
+
"eval_weighted_sensitivity": 0.8062360801781737,
|
234 |
+
"eval_weighted_specificity": 0.8594586246285271,
|
235 |
+
"step": 2016
|
236 |
+
},
|
237 |
+
{
|
238 |
+
"epoch": 10.0,
|
239 |
+
"eval_accuracy": 0.8173719376391982,
|
240 |
+
"eval_f1_macro": 0.6554404393882288,
|
241 |
+
"eval_f1_micro": 0.8173719376391981,
|
242 |
+
"eval_f1_weighted": 0.8137262031233367,
|
243 |
+
"eval_loss": 1.2516238689422607,
|
244 |
+
"eval_macro_fpr": 0.08727001865560947,
|
245 |
+
"eval_macro_sensitivity": 0.649889194006841,
|
246 |
+
"eval_macro_specificity": 0.9218678293458737,
|
247 |
+
"eval_precision": 0.8106611480963274,
|
248 |
+
"eval_precision_macro": 0.6620602008894919,
|
249 |
+
"eval_recall": 0.8173719376391982,
|
250 |
+
"eval_recall_macro": 0.649889194006841,
|
251 |
+
"eval_runtime": 34.0534,
|
252 |
+
"eval_samples_per_second": 13.185,
|
253 |
+
"eval_steps_per_second": 1.674,
|
254 |
+
"eval_weighted_fpr": 0.06931530008453085,
|
255 |
+
"eval_weighted_sensitivity": 0.8173719376391982,
|
256 |
+
"eval_weighted_specificity": 0.8700993797442969,
|
257 |
+
"step": 2240
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 11.0,
|
261 |
+
"eval_accuracy": 0.8262806236080178,
|
262 |
+
"eval_f1_macro": 0.6764229637898442,
|
263 |
+
"eval_f1_micro": 0.8262806236080178,
|
264 |
+
"eval_f1_weighted": 0.819522907157828,
|
265 |
+
"eval_loss": 1.3822201490402222,
|
266 |
+
"eval_macro_fpr": 0.08329902302757192,
|
267 |
+
"eval_macro_sensitivity": 0.669645007880302,
|
268 |
+
"eval_macro_specificity": 0.9243474796823474,
|
269 |
+
"eval_precision": 0.8204585488197453,
|
270 |
+
"eval_precision_macro": 0.7084622597280825,
|
271 |
+
"eval_recall": 0.8262806236080178,
|
272 |
+
"eval_recall_macro": 0.669645007880302,
|
273 |
+
"eval_runtime": 31.5236,
|
274 |
+
"eval_samples_per_second": 14.243,
|
275 |
+
"eval_steps_per_second": 1.808,
|
276 |
+
"eval_weighted_fpr": 0.0654911838790932,
|
277 |
+
"eval_weighted_sensitivity": 0.8262806236080178,
|
278 |
+
"eval_weighted_specificity": 0.8711092951213717,
|
279 |
+
"step": 2464
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 11.160714285714286,
|
283 |
+
"grad_norm": 2.0276401042938232,
|
284 |
+
"learning_rate": 3.1443452380952384e-05,
|
285 |
+
"loss": 0.055,
|
286 |
+
"step": 2500
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 12.0,
|
290 |
+
"eval_accuracy": 0.8017817371937639,
|
291 |
+
"eval_f1_macro": 0.6399377531674577,
|
292 |
+
"eval_f1_micro": 0.8017817371937639,
|
293 |
+
"eval_f1_weighted": 0.806834856085664,
|
294 |
+
"eval_loss": 1.4574010372161865,
|
295 |
+
"eval_macro_fpr": 0.0883315306539799,
|
296 |
+
"eval_macro_sensitivity": 0.6443222252045782,
|
297 |
+
"eval_macro_specificity": 0.9215573710717683,
|
298 |
+
"eval_precision": 0.8127232508105453,
|
299 |
+
"eval_precision_macro": 0.6368985338108845,
|
300 |
+
"eval_recall": 0.8017817371937639,
|
301 |
+
"eval_recall_macro": 0.6443222252045782,
|
302 |
+
"eval_runtime": 35.9686,
|
303 |
+
"eval_samples_per_second": 12.483,
|
304 |
+
"eval_steps_per_second": 1.585,
|
305 |
+
"eval_weighted_fpr": 0.07613344739093242,
|
306 |
+
"eval_weighted_sensitivity": 0.8017817371937639,
|
307 |
+
"eval_weighted_specificity": 0.884447747093309,
|
308 |
+
"step": 2688
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 13.0,
|
312 |
+
"eval_accuracy": 0.7884187082405345,
|
313 |
+
"eval_f1_macro": 0.6042489773120808,
|
314 |
+
"eval_f1_micro": 0.7884187082405345,
|
315 |
+
"eval_f1_weighted": 0.7831300624317719,
|
316 |
+
"eval_loss": 1.6634190082550049,
|
317 |
+
"eval_macro_fpr": 0.10018097159570645,
|
318 |
+
"eval_macro_sensitivity": 0.6042144130379424,
|
319 |
+
"eval_macro_specificity": 0.9125690965302461,
|
320 |
+
"eval_precision": 0.7810481910770134,
|
321 |
+
"eval_precision_macro": 0.6089900703871292,
|
322 |
+
"eval_recall": 0.7884187082405345,
|
323 |
+
"eval_recall_macro": 0.6042144130379424,
|
324 |
+
"eval_runtime": 29.9493,
|
325 |
+
"eval_samples_per_second": 14.992,
|
326 |
+
"eval_steps_per_second": 1.903,
|
327 |
+
"eval_weighted_fpr": 0.08210890233362143,
|
328 |
+
"eval_weighted_sensitivity": 0.7884187082405345,
|
329 |
+
"eval_weighted_specificity": 0.8618576778804495,
|
330 |
+
"step": 2912
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 13.392857142857142,
|
334 |
+
"grad_norm": 0.00395896565169096,
|
335 |
+
"learning_rate": 2.773809523809524e-05,
|
336 |
+
"loss": 0.0431,
|
337 |
+
"step": 3000
|
338 |
+
},
|
339 |
+
{
|
340 |
+
"epoch": 14.0,
|
341 |
+
"eval_accuracy": 0.8285077951002228,
|
342 |
+
"eval_f1_macro": 0.6381695277501472,
|
343 |
+
"eval_f1_micro": 0.8285077951002228,
|
344 |
+
"eval_f1_weighted": 0.8165889625025802,
|
345 |
+
"eval_loss": 1.5085251331329346,
|
346 |
+
"eval_macro_fpr": 0.0849606439034177,
|
347 |
+
"eval_macro_sensitivity": 0.6367413058589528,
|
348 |
+
"eval_macro_specificity": 0.9229424414178813,
|
349 |
+
"eval_precision": 0.8077049262727573,
|
350 |
+
"eval_precision_macro": 0.6475977473768657,
|
351 |
+
"eval_recall": 0.8285077951002228,
|
352 |
+
"eval_recall_macro": 0.6367413058589528,
|
353 |
+
"eval_runtime": 34.3243,
|
354 |
+
"eval_samples_per_second": 13.081,
|
355 |
+
"eval_steps_per_second": 1.661,
|
356 |
+
"eval_weighted_fpr": 0.06454316848281642,
|
357 |
+
"eval_weighted_sensitivity": 0.8285077951002228,
|
358 |
+
"eval_weighted_specificity": 0.8632619705713027,
|
359 |
+
"step": 3136
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 15.0,
|
363 |
+
"eval_accuracy": 0.8106904231625836,
|
364 |
+
"eval_f1_macro": 0.6229362972793631,
|
365 |
+
"eval_f1_micro": 0.8106904231625834,
|
366 |
+
"eval_f1_weighted": 0.8013640427817921,
|
367 |
+
"eval_loss": 1.641113042831421,
|
368 |
+
"eval_macro_fpr": 0.09140135508148657,
|
369 |
+
"eval_macro_sensitivity": 0.6261966703143174,
|
370 |
+
"eval_macro_specificity": 0.9183149213614681,
|
371 |
+
"eval_precision": 0.7935778228683108,
|
372 |
+
"eval_precision_macro": 0.6242729514731414,
|
373 |
+
"eval_recall": 0.8106904231625836,
|
374 |
+
"eval_recall_macro": 0.6261966703143174,
|
375 |
+
"eval_runtime": 35.8672,
|
376 |
+
"eval_samples_per_second": 12.518,
|
377 |
+
"eval_steps_per_second": 1.589,
|
378 |
+
"eval_weighted_fpr": 0.07221750212404418,
|
379 |
+
"eval_weighted_sensitivity": 0.8106904231625836,
|
380 |
+
"eval_weighted_specificity": 0.8625692622832889,
|
381 |
+
"step": 3360
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 15.625,
|
385 |
+
"grad_norm": 0.09424740076065063,
|
386 |
+
"learning_rate": 2.4017857142857146e-05,
|
387 |
+
"loss": 0.0135,
|
388 |
+
"step": 3500
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 16.0,
|
392 |
+
"eval_accuracy": 0.8062360801781737,
|
393 |
+
"eval_f1_macro": 0.6221390381080233,
|
394 |
+
"eval_f1_micro": 0.8062360801781738,
|
395 |
+
"eval_f1_weighted": 0.7988403583111869,
|
396 |
+
"eval_loss": 1.7482596635818481,
|
397 |
+
"eval_macro_fpr": 0.09232876528599361,
|
398 |
+
"eval_macro_sensitivity": 0.6270887962064433,
|
399 |
+
"eval_macro_specificity": 0.9177392895526978,
|
400 |
+
"eval_precision": 0.7924673420177055,
|
401 |
+
"eval_precision_macro": 0.6200610055562886,
|
402 |
+
"eval_recall": 0.8062360801781737,
|
403 |
+
"eval_recall_macro": 0.6270887962064433,
|
404 |
+
"eval_runtime": 34.8393,
|
405 |
+
"eval_samples_per_second": 12.888,
|
406 |
+
"eval_steps_per_second": 1.636,
|
407 |
+
"eval_weighted_fpr": 0.0741687979539642,
|
408 |
+
"eval_weighted_sensitivity": 0.8062360801781737,
|
409 |
+
"eval_weighted_specificity": 0.8647210780326176,
|
410 |
+
"step": 3584
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 17.0,
|
414 |
+
"eval_accuracy": 0.7973273942093542,
|
415 |
+
"eval_f1_macro": 0.6201227671521832,
|
416 |
+
"eval_f1_micro": 0.7973273942093542,
|
417 |
+
"eval_f1_weighted": 0.7932689879372665,
|
418 |
+
"eval_loss": 1.7232707738876343,
|
419 |
+
"eval_macro_fpr": 0.09423113155723643,
|
420 |
+
"eval_macro_sensitivity": 0.6263480822304351,
|
421 |
+
"eval_macro_specificity": 0.9163848671929355,
|
422 |
+
"eval_precision": 0.789684456744836,
|
423 |
+
"eval_precision_macro": 0.6147620506535948,
|
424 |
+
"eval_recall": 0.7973273942093542,
|
425 |
+
"eval_recall_macro": 0.6263480822304351,
|
426 |
+
"eval_runtime": 36.3399,
|
427 |
+
"eval_samples_per_second": 12.356,
|
428 |
+
"eval_steps_per_second": 1.569,
|
429 |
+
"eval_weighted_fpr": 0.07811158798283262,
|
430 |
+
"eval_weighted_sensitivity": 0.7973273942093542,
|
431 |
+
"eval_weighted_specificity": 0.8682120745623881,
|
432 |
+
"step": 3808
|
433 |
+
},
|
434 |
+
{
|
435 |
+
"epoch": 17.857142857142858,
|
436 |
+
"grad_norm": 0.004108293913304806,
|
437 |
+
"learning_rate": 2.029761904761905e-05,
|
438 |
+
"loss": 0.0066,
|
439 |
+
"step": 4000
|
440 |
+
},
|
441 |
+
{
|
442 |
+
"epoch": 18.0,
|
443 |
+
"eval_accuracy": 0.8240534521158129,
|
444 |
+
"eval_f1_macro": 0.639053660901891,
|
445 |
+
"eval_f1_micro": 0.8240534521158129,
|
446 |
+
"eval_f1_weighted": 0.8115429423166062,
|
447 |
+
"eval_loss": 1.6456769704818726,
|
448 |
+
"eval_macro_fpr": 0.08791315651117829,
|
449 |
+
"eval_macro_sensitivity": 0.6388214991156167,
|
450 |
+
"eval_macro_specificity": 0.9190529442926627,
|
451 |
+
"eval_precision": 0.8041665668829977,
|
452 |
+
"eval_precision_macro": 0.6515456989247311,
|
453 |
+
"eval_recall": 0.8240534521158129,
|
454 |
+
"eval_recall_macro": 0.6388214991156167,
|
455 |
+
"eval_runtime": 33.8794,
|
456 |
+
"eval_samples_per_second": 13.253,
|
457 |
+
"eval_steps_per_second": 1.682,
|
458 |
+
"eval_weighted_fpr": 0.06644238856181665,
|
459 |
+
"eval_weighted_sensitivity": 0.8240534521158129,
|
460 |
+
"eval_weighted_specificity": 0.8521583250548381,
|
461 |
+
"step": 4032
|
462 |
+
}
|
463 |
+
],
|
464 |
+
"logging_steps": 500,
|
465 |
+
"max_steps": 6720,
|
466 |
+
"num_input_tokens_seen": 0,
|
467 |
+
"num_train_epochs": 30,
|
468 |
+
"save_steps": 500,
|
469 |
+
"total_flos": 8487062602776576.0,
|
470 |
+
"train_batch_size": 8,
|
471 |
+
"trial_name": null,
|
472 |
+
"trial_params": null
|
473 |
+
}
|
training_checkpoints/checkpoint-4032/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8124b12bb2d716832b587bea11323810e8fd3f54cf265a6935c0b9d891cdd1f6
|
3 |
+
size 5048
|
training_checkpoints/checkpoint-4032/vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|