avijit commited on
Commit
e247457
·
verified ·
1 Parent(s): 7e17e34

Upload folder using huggingface_hub

Browse files
added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<mask>": 50000
3
+ }
config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/home/avijit/playground/sdoh/RoBERTa-base-PM-M3-Voc-distill-align",
3
+ "architectures": [
4
+ "RobertaForMultiLabelClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "gradient_checkpointing": false,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "id2label": {
15
+ "0": "LABEL_0",
16
+ "1": "LABEL_1",
17
+ "2": "LABEL_2",
18
+ "3": "LABEL_3",
19
+ "4": "LABEL_4",
20
+ "5": "LABEL_5",
21
+ "6": "LABEL_6",
22
+ "7": "LABEL_7",
23
+ "8": "LABEL_8",
24
+ "9": "LABEL_9",
25
+ "10": "LABEL_10",
26
+ "11": "LABEL_11"
27
+ },
28
+ "initializer_range": 0.02,
29
+ "intermediate_size": 3072,
30
+ "label2id": {
31
+ "LABEL_0": 0,
32
+ "LABEL_1": 1,
33
+ "LABEL_10": 10,
34
+ "LABEL_11": 11,
35
+ "LABEL_2": 2,
36
+ "LABEL_3": 3,
37
+ "LABEL_4": 4,
38
+ "LABEL_5": 5,
39
+ "LABEL_6": 6,
40
+ "LABEL_7": 7,
41
+ "LABEL_8": 8,
42
+ "LABEL_9": 9
43
+ },
44
+ "layer_norm_eps": 1e-05,
45
+ "max_position_embeddings": 514,
46
+ "model_type": "roberta",
47
+ "num_attention_heads": 12,
48
+ "num_hidden_layers": 12,
49
+ "pad_token_id": 1,
50
+ "position_embedding_type": "absolute",
51
+ "problem_type": "multi_label_classification",
52
+ "torch_dtype": "float32",
53
+ "transformers_version": "4.39.1",
54
+ "type_vocab_size": 1,
55
+ "use_cache": true,
56
+ "vocab_size": 50008
57
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a25b9839d4cd8ce15c918fe0fb08e244cc83ad097c276985e1e185efe6ad7e3
3
+ size 497854080
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9ce9c83507389cbceb47d23b3b191b2a8fbf14b7dfeb686cc702b938985cc37e
3
+ size 995827962
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3533dad152f2261c2511c7e24ec09386552bab04d4a8c5104206019a9b5e4721
3
+ size 14180
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd73f99bfa1e38f01efa913f5246efefdf17b845643f2e7b43c26056675c0848
3
+ size 1064
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": true,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": true,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "50000": {
37
+ "content": "<mask>",
38
+ "lstrip": true,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "bos_token": "<s>",
46
+ "clean_up_tokenization_spaces": true,
47
+ "cls_token": "<s>",
48
+ "eos_token": "</s>",
49
+ "errors": "replace",
50
+ "mask_token": "<mask>",
51
+ "model_max_length": 1000000000000000019884624838656,
52
+ "pad_token": "<pad>",
53
+ "sep_token": "</s>",
54
+ "tokenizer_class": "RobertaTokenizer",
55
+ "trim_offsets": true,
56
+ "unk_token": "<unk>"
57
+ }
trainer_state.json ADDED
@@ -0,0 +1,1199 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9129580069246931,
3
+ "best_model_checkpoint": "./saved_models/cliroberta_sbdh_gpt4_v2_0/checkpoint-912",
4
+ "epoch": 38.0,
5
+ "eval_steps": 500,
6
+ "global_step": 912,
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": 0.9901933670043945,
14
+ "learning_rate": 2.5e-06,
15
+ "loss": 0.575,
16
+ "step": 24
17
+ },
18
+ {
19
+ "epoch": 1.0,
20
+ "eval_acc_macro": 0.1223753452493468,
21
+ "eval_acc_micro": 0.14617249740214502,
22
+ "eval_auc_macro": 0.6179498256291489,
23
+ "eval_auc_micro": 0.6115176410375945,
24
+ "eval_f1_at_5": 0.24653561710044558,
25
+ "eval_f1_at_8": 0.23250745850924381,
26
+ "eval_f1_macro": 0.20626681869979227,
27
+ "eval_f1_micro": 0.25506195225142775,
28
+ "eval_loss": 0.4440538287162781,
29
+ "eval_prec_at_5": 0.15890410958904108,
30
+ "eval_prec_at_8": 0.13684360730593606,
31
+ "eval_prec_macro": 0.16142693103921257,
32
+ "eval_prec_micro": 0.15594974131559208,
33
+ "eval_rec_at_5": 0.5496575342465754,
34
+ "eval_rec_at_8": 0.772640791476408,
35
+ "eval_rec_macro": 0.598680624031903,
36
+ "eval_rec_micro": 0.6998341625206717,
37
+ "eval_runtime": 3.2806,
38
+ "eval_samples_per_second": 267.022,
39
+ "eval_steps_per_second": 33.53,
40
+ "step": 24
41
+ },
42
+ {
43
+ "epoch": 2.0,
44
+ "grad_norm": 0.45090538263320923,
45
+ "learning_rate": 5e-06,
46
+ "loss": 0.416,
47
+ "step": 48
48
+ },
49
+ {
50
+ "epoch": 2.0,
51
+ "eval_acc_macro": 0.12473782113690822,
52
+ "eval_acc_micro": 0.19038767923525782,
53
+ "eval_auc_macro": 0.7812955931842048,
54
+ "eval_auc_micro": 0.7120923874787535,
55
+ "eval_f1_at_5": 0.28976519882341023,
56
+ "eval_f1_at_8": 0.26129064546572484,
57
+ "eval_f1_macro": 0.20647502666276477,
58
+ "eval_f1_micro": 0.3198750836493277,
59
+ "eval_loss": 0.35849881172180176,
60
+ "eval_prec_at_5": 0.1860730593607306,
61
+ "eval_prec_at_8": 0.15368150684931506,
62
+ "eval_prec_macro": 0.2441853686080976,
63
+ "eval_prec_micro": 0.2187976808056082,
64
+ "eval_rec_at_5": 0.654490106544901,
65
+ "eval_rec_at_8": 0.8715753424657534,
66
+ "eval_rec_macro": 0.4255708847629977,
67
+ "eval_rec_micro": 0.5945273631840303,
68
+ "eval_runtime": 3.1796,
69
+ "eval_samples_per_second": 275.508,
70
+ "eval_steps_per_second": 34.596,
71
+ "step": 48
72
+ },
73
+ {
74
+ "epoch": 3.0,
75
+ "grad_norm": 0.27397865056991577,
76
+ "learning_rate": 7.500000000000001e-06,
77
+ "loss": 0.3638,
78
+ "step": 72
79
+ },
80
+ {
81
+ "epoch": 3.0,
82
+ "eval_acc_macro": 0.08130874759470162,
83
+ "eval_acc_micro": 0.24049000644743773,
84
+ "eval_auc_macro": 0.8877837878947817,
85
+ "eval_auc_micro": 0.7666590840482023,
86
+ "eval_f1_at_5": 0.336822490351584,
87
+ "eval_f1_at_8": 0.2698854516748229,
88
+ "eval_f1_macro": 0.10904515062726224,
89
+ "eval_f1_micro": 0.3877338877338474,
90
+ "eval_loss": 0.3340158760547638,
91
+ "eval_prec_at_5": 0.21575342465753425,
92
+ "eval_prec_at_8": 0.158675799086758,
93
+ "eval_prec_macro": 0.08635061077641497,
94
+ "eval_prec_micro": 0.5194986072422675,
95
+ "eval_rec_at_5": 0.7675038051750379,
96
+ "eval_rec_at_8": 0.9022070015220699,
97
+ "eval_rec_macro": 0.147953216374198,
98
+ "eval_rec_micro": 0.309286898839112,
99
+ "eval_runtime": 3.1991,
100
+ "eval_samples_per_second": 273.823,
101
+ "eval_steps_per_second": 34.384,
102
+ "step": 72
103
+ },
104
+ {
105
+ "epoch": 4.0,
106
+ "grad_norm": 0.2640039920806885,
107
+ "learning_rate": 1e-05,
108
+ "loss": 0.3328,
109
+ "step": 96
110
+ },
111
+ {
112
+ "epoch": 4.0,
113
+ "eval_acc_macro": 0.2356290306850708,
114
+ "eval_acc_micro": 0.36202531645567326,
115
+ "eval_auc_macro": 0.9194506958206632,
116
+ "eval_auc_micro": 0.8662944233628048,
117
+ "eval_f1_at_5": 0.3860939867346734,
118
+ "eval_f1_at_8": 0.2863186309798368,
119
+ "eval_f1_macro": 0.31624180053900214,
120
+ "eval_f1_micro": 0.5315985130111031,
121
+ "eval_loss": 0.29880067706108093,
122
+ "eval_prec_at_5": 0.24703196347031964,
123
+ "eval_prec_at_8": 0.168236301369863,
124
+ "eval_prec_macro": 0.38775967404251327,
125
+ "eval_prec_micro": 0.6046511627906337,
126
+ "eval_rec_at_5": 0.8833713850837139,
127
+ "eval_rec_at_8": 0.9604261796042617,
128
+ "eval_rec_macro": 0.31299638776931427,
129
+ "eval_rec_micro": 0.4742951907130618,
130
+ "eval_runtime": 3.2144,
131
+ "eval_samples_per_second": 272.527,
132
+ "eval_steps_per_second": 34.221,
133
+ "step": 96
134
+ },
135
+ {
136
+ "epoch": 5.0,
137
+ "grad_norm": 0.29883790016174316,
138
+ "learning_rate": 9.722222222222223e-06,
139
+ "loss": 0.2961,
140
+ "step": 120
141
+ },
142
+ {
143
+ "epoch": 5.0,
144
+ "eval_acc_macro": 0.45969839854913025,
145
+ "eval_acc_micro": 0.5302557704304098,
146
+ "eval_auc_macro": 0.9400078444800295,
147
+ "eval_auc_micro": 0.9362775366665492,
148
+ "eval_f1_at_5": 0.41206764411360836,
149
+ "eval_f1_at_8": 0.2904319455576605,
150
+ "eval_f1_macro": 0.5639799795799724,
151
+ "eval_f1_micro": 0.6930289441499639,
152
+ "eval_loss": 0.25566670298576355,
153
+ "eval_prec_at_5": 0.26324200913242013,
154
+ "eval_prec_at_8": 0.170662100456621,
155
+ "eval_prec_macro": 0.5869010270714891,
156
+ "eval_prec_micro": 0.6816359262228804,
157
+ "eval_rec_at_5": 0.9480593607305936,
158
+ "eval_rec_at_8": 0.9739345509893454,
159
+ "eval_rec_macro": 0.5895666448012687,
160
+ "eval_rec_micro": 0.7048092868987806,
161
+ "eval_runtime": 3.2031,
162
+ "eval_samples_per_second": 273.486,
163
+ "eval_steps_per_second": 34.342,
164
+ "step": 120
165
+ },
166
+ {
167
+ "epoch": 6.0,
168
+ "grad_norm": 0.3095230162143707,
169
+ "learning_rate": 9.444444444444445e-06,
170
+ "loss": 0.2582,
171
+ "step": 144
172
+ },
173
+ {
174
+ "epoch": 6.0,
175
+ "eval_acc_macro": 0.5571172225508884,
176
+ "eval_acc_micro": 0.599110546378615,
177
+ "eval_auc_macro": 0.957377398100574,
178
+ "eval_auc_micro": 0.9585021379241767,
179
+ "eval_f1_at_5": 0.41988358958639926,
180
+ "eval_f1_at_8": 0.2904319455576605,
181
+ "eval_f1_macro": 0.6798913628949355,
182
+ "eval_f1_micro": 0.7493047278505561,
183
+ "eval_loss": 0.22171612083911896,
184
+ "eval_prec_at_5": 0.2684931506849315,
185
+ "eval_prec_at_8": 0.170662100456621,
186
+ "eval_prec_macro": 0.7360995034850415,
187
+ "eval_prec_micro": 0.7192982456139801,
188
+ "eval_rec_at_5": 0.9627092846270927,
189
+ "eval_rec_at_8": 0.9739345509893454,
190
+ "eval_rec_macro": 0.7032528362779372,
191
+ "eval_rec_micro": 0.7819237147594708,
192
+ "eval_runtime": 3.2002,
193
+ "eval_samples_per_second": 273.729,
194
+ "eval_steps_per_second": 34.372,
195
+ "step": 144
196
+ },
197
+ {
198
+ "epoch": 7.0,
199
+ "grad_norm": 0.4028068780899048,
200
+ "learning_rate": 9.166666666666666e-06,
201
+ "loss": 0.2275,
202
+ "step": 168
203
+ },
204
+ {
205
+ "epoch": 7.0,
206
+ "eval_acc_macro": 0.6549888166459882,
207
+ "eval_acc_micro": 0.6949270326615222,
208
+ "eval_auc_macro": 0.9666862869707499,
209
+ "eval_auc_micro": 0.9705963698236378,
210
+ "eval_f1_at_5": 0.421956752081853,
211
+ "eval_f1_at_8": 0.2911364149806836,
212
+ "eval_f1_macro": 0.769457395477212,
213
+ "eval_f1_micro": 0.8200082000819336,
214
+ "eval_loss": 0.19522725045681,
215
+ "eval_prec_at_5": 0.2698630136986302,
216
+ "eval_prec_at_8": 0.17109018264840184,
217
+ "eval_prec_macro": 0.8425529900891876,
218
+ "eval_prec_micro": 0.8110300081102343,
219
+ "eval_rec_at_5": 0.9668949771689498,
220
+ "eval_rec_at_8": 0.9758371385083714,
221
+ "eval_rec_macro": 0.7739718206004228,
222
+ "eval_rec_micro": 0.8291873963515066,
223
+ "eval_runtime": 3.1872,
224
+ "eval_samples_per_second": 274.847,
225
+ "eval_steps_per_second": 34.513,
226
+ "step": 168
227
+ },
228
+ {
229
+ "epoch": 8.0,
230
+ "grad_norm": 0.3119203448295593,
231
+ "learning_rate": 8.888888888888888e-06,
232
+ "loss": 0.2028,
233
+ "step": 192
234
+ },
235
+ {
236
+ "epoch": 8.0,
237
+ "eval_acc_macro": 0.6972915753471085,
238
+ "eval_acc_micro": 0.7357664233576106,
239
+ "eval_auc_macro": 0.9740916521582301,
240
+ "eval_auc_micro": 0.9777448811533707,
241
+ "eval_f1_at_5": 0.4238663640920515,
242
+ "eval_f1_at_8": 0.2915919235644548,
243
+ "eval_f1_macro": 0.8073127904534504,
244
+ "eval_f1_micro": 0.8477712363329817,
245
+ "eval_loss": 0.17399446666240692,
246
+ "eval_prec_at_5": 0.2712328767123288,
247
+ "eval_prec_at_8": 0.1713755707762557,
248
+ "eval_prec_macro": 0.8738560982440289,
249
+ "eval_prec_micro": 0.8600682593855921,
250
+ "eval_rec_at_5": 0.9693683409436834,
251
+ "eval_rec_at_8": 0.9767884322678843,
252
+ "eval_rec_macro": 0.7867760719656439,
253
+ "eval_rec_micro": 0.8358208955223188,
254
+ "eval_runtime": 3.1788,
255
+ "eval_samples_per_second": 275.577,
256
+ "eval_steps_per_second": 34.604,
257
+ "step": 192
258
+ },
259
+ {
260
+ "epoch": 9.0,
261
+ "grad_norm": 0.3134922683238983,
262
+ "learning_rate": 8.611111111111112e-06,
263
+ "loss": 0.1824,
264
+ "step": 216
265
+ },
266
+ {
267
+ "epoch": 9.0,
268
+ "eval_acc_macro": 0.7304201087212143,
269
+ "eval_acc_micro": 0.7586206896551179,
270
+ "eval_auc_macro": 0.9803240639791783,
271
+ "eval_auc_micro": 0.9830083410585158,
272
+ "eval_f1_at_5": 0.42521790481745503,
273
+ "eval_f1_at_8": 0.2922793659426448,
274
+ "eval_f1_macro": 0.835259929842726,
275
+ "eval_f1_micro": 0.8627450980391451,
276
+ "eval_loss": 0.1565641313791275,
277
+ "eval_prec_at_5": 0.2721461187214612,
278
+ "eval_prec_at_8": 0.17180365296803654,
279
+ "eval_prec_macro": 0.8537779731664906,
280
+ "eval_prec_micro": 0.8502415458936513,
281
+ "eval_rec_at_5": 0.9718417047184171,
282
+ "eval_rec_at_8": 0.978310502283105,
283
+ "eval_rec_macro": 0.838195231781537,
284
+ "eval_rec_micro": 0.8756218905471911,
285
+ "eval_runtime": 3.2483,
286
+ "eval_samples_per_second": 269.681,
287
+ "eval_steps_per_second": 33.864,
288
+ "step": 216
289
+ },
290
+ {
291
+ "epoch": 10.0,
292
+ "grad_norm": 0.4643474221229553,
293
+ "learning_rate": 8.333333333333334e-06,
294
+ "loss": 0.1651,
295
+ "step": 240
296
+ },
297
+ {
298
+ "epoch": 10.0,
299
+ "eval_acc_macro": 0.7537121442185194,
300
+ "eval_acc_micro": 0.7786705624542893,
301
+ "eval_auc_macro": 0.9820077423929909,
302
+ "eval_auc_micro": 0.9843423829345286,
303
+ "eval_f1_at_5": 0.4252361135724823,
304
+ "eval_f1_at_8": 0.2915919235644548,
305
+ "eval_f1_macro": 0.852437891908891,
306
+ "eval_f1_micro": 0.8755646817247741,
307
+ "eval_loss": 0.14288866519927979,
308
+ "eval_prec_at_5": 0.2721461187214612,
309
+ "eval_prec_at_8": 0.1713755707762557,
310
+ "eval_prec_macro": 0.8628168198092824,
311
+ "eval_prec_micro": 0.8673718470300352,
312
+ "eval_rec_at_5": 0.9720319634703196,
313
+ "eval_rec_at_8": 0.9767884322678843,
314
+ "eval_rec_macro": 0.8531543520415448,
315
+ "eval_rec_micro": 0.8839137645107061,
316
+ "eval_runtime": 3.1569,
317
+ "eval_samples_per_second": 277.491,
318
+ "eval_steps_per_second": 34.845,
319
+ "step": 240
320
+ },
321
+ {
322
+ "epoch": 11.0,
323
+ "grad_norm": 0.41132673621177673,
324
+ "learning_rate": 8.055555555555557e-06,
325
+ "loss": 0.1507,
326
+ "step": 264
327
+ },
328
+ {
329
+ "epoch": 11.0,
330
+ "eval_acc_macro": 0.7725786729049252,
331
+ "eval_acc_micro": 0.7965465465464867,
332
+ "eval_auc_macro": 0.9845391365423227,
333
+ "eval_auc_micro": 0.9861133832235769,
334
+ "eval_f1_at_5": 0.42516324511829706,
335
+ "eval_f1_at_8": 0.2915919235644548,
336
+ "eval_f1_macro": 0.8654438329084839,
337
+ "eval_f1_micro": 0.8867530296697962,
338
+ "eval_loss": 0.13122519850730896,
339
+ "eval_prec_at_5": 0.2721461187214612,
340
+ "eval_prec_at_8": 0.1713755707762557,
341
+ "eval_prec_macro": 0.890999273408568,
342
+ "eval_prec_micro": 0.8938500421229238,
343
+ "eval_rec_at_5": 0.9712709284627093,
344
+ "eval_rec_at_8": 0.9767884322678843,
345
+ "eval_rec_macro": 0.8506625538285683,
346
+ "eval_rec_micro": 0.8797678275289486,
347
+ "eval_runtime": 3.1904,
348
+ "eval_samples_per_second": 274.571,
349
+ "eval_steps_per_second": 34.478,
350
+ "step": 264
351
+ },
352
+ {
353
+ "epoch": 12.0,
354
+ "grad_norm": 0.5822508335113525,
355
+ "learning_rate": 7.77777777777778e-06,
356
+ "loss": 0.1381,
357
+ "step": 288
358
+ },
359
+ {
360
+ "epoch": 12.0,
361
+ "eval_acc_macro": 0.7941812566844084,
362
+ "eval_acc_micro": 0.8098434004473669,
363
+ "eval_auc_macro": 0.9868855804701235,
364
+ "eval_auc_micro": 0.9890811185137426,
365
+ "eval_f1_at_5": 0.4261995886925304,
366
+ "eval_f1_at_8": 0.29205588274802374,
367
+ "eval_f1_macro": 0.8806793746749714,
368
+ "eval_f1_micro": 0.8949320148330534,
369
+ "eval_loss": 0.12080539762973785,
370
+ "eval_prec_at_5": 0.27283105022831056,
371
+ "eval_prec_at_8": 0.1716609589041096,
372
+ "eval_prec_macro": 0.8843316154214099,
373
+ "eval_prec_micro": 0.8894348894348165,
374
+ "eval_rec_at_5": 0.9733637747336377,
375
+ "eval_rec_at_8": 0.9779299847792998,
376
+ "eval_rec_macro": 0.8816174783898175,
377
+ "eval_rec_micro": 0.9004975124377362,
378
+ "eval_runtime": 3.278,
379
+ "eval_samples_per_second": 267.232,
380
+ "eval_steps_per_second": 33.557,
381
+ "step": 288
382
+ },
383
+ {
384
+ "epoch": 13.0,
385
+ "grad_norm": 0.5692580342292786,
386
+ "learning_rate": 7.500000000000001e-06,
387
+ "loss": 0.1278,
388
+ "step": 312
389
+ },
390
+ {
391
+ "epoch": 13.0,
392
+ "eval_acc_macro": 0.810678005842559,
393
+ "eval_acc_micro": 0.8233082706766298,
394
+ "eval_auc_macro": 0.9875718328324142,
395
+ "eval_auc_micro": 0.989483950688566,
396
+ "eval_f1_at_5": 0.4271994046239887,
397
+ "eval_f1_at_8": 0.29182390543479947,
398
+ "eval_f1_macro": 0.8918987653168854,
399
+ "eval_f1_micro": 0.90309278350508,
400
+ "eval_loss": 0.11230690032243729,
401
+ "eval_prec_at_5": 0.27351598173515984,
402
+ "eval_prec_at_8": 0.17151826484018265,
403
+ "eval_prec_macro": 0.8939579787040955,
404
+ "eval_prec_micro": 0.8982772764560378,
405
+ "eval_rec_at_5": 0.975076103500761,
406
+ "eval_rec_at_8": 0.977359208523592,
407
+ "eval_rec_macro": 0.8924277193620916,
408
+ "eval_rec_micro": 0.9079601990048998,
409
+ "eval_runtime": 3.2773,
410
+ "eval_samples_per_second": 267.291,
411
+ "eval_steps_per_second": 33.564,
412
+ "step": 312
413
+ },
414
+ {
415
+ "epoch": 14.0,
416
+ "grad_norm": 0.4341146647930145,
417
+ "learning_rate": 7.222222222222223e-06,
418
+ "loss": 0.1181,
419
+ "step": 336
420
+ },
421
+ {
422
+ "epoch": 14.0,
423
+ "eval_acc_macro": 0.8114177963739652,
424
+ "eval_acc_micro": 0.8276661514682513,
425
+ "eval_auc_macro": 0.9882489665360898,
426
+ "eval_auc_micro": 0.9896063774543715,
427
+ "eval_f1_at_5": 0.4275326584009282,
428
+ "eval_f1_at_8": 0.2915919235644548,
429
+ "eval_f1_macro": 0.8910138454396187,
430
+ "eval_f1_micro": 0.9057082452430524,
431
+ "eval_loss": 0.10568831115961075,
432
+ "eval_prec_at_5": 0.27374429223744295,
433
+ "eval_prec_at_8": 0.1713755707762557,
434
+ "eval_prec_macro": 0.9229102543764114,
435
+ "eval_prec_micro": 0.924072476272569,
436
+ "eval_rec_at_5": 0.9756468797564688,
437
+ "eval_rec_at_8": 0.9767884322678843,
438
+ "eval_rec_macro": 0.8685981119086549,
439
+ "eval_rec_micro": 0.8880597014924636,
440
+ "eval_runtime": 3.2679,
441
+ "eval_samples_per_second": 268.059,
442
+ "eval_steps_per_second": 33.66,
443
+ "step": 336
444
+ },
445
+ {
446
+ "epoch": 15.0,
447
+ "grad_norm": 0.3292841613292694,
448
+ "learning_rate": 6.944444444444445e-06,
449
+ "loss": 0.1101,
450
+ "step": 360
451
+ },
452
+ {
453
+ "epoch": 15.0,
454
+ "eval_acc_macro": 0.8196380626953511,
455
+ "eval_acc_micro": 0.8308880308879667,
456
+ "eval_auc_macro": 0.9889504342095332,
457
+ "eval_auc_micro": 0.9907443048387263,
458
+ "eval_f1_at_5": 0.4271994046239887,
459
+ "eval_f1_at_8": 0.29182390543479947,
460
+ "eval_f1_macro": 0.8973412777794657,
461
+ "eval_f1_micro": 0.9076339097426479,
462
+ "eval_loss": 0.0997268408536911,
463
+ "eval_prec_at_5": 0.27351598173515984,
464
+ "eval_prec_at_8": 0.17151826484018265,
465
+ "eval_prec_macro": 0.9202983062064826,
466
+ "eval_prec_micro": 0.923605150214513,
467
+ "eval_rec_at_5": 0.975076103500761,
468
+ "eval_rec_at_8": 0.977359208523592,
469
+ "eval_rec_macro": 0.8785932935910168,
470
+ "eval_rec_micro": 0.8922056384742212,
471
+ "eval_runtime": 3.2426,
472
+ "eval_samples_per_second": 270.154,
473
+ "eval_steps_per_second": 33.923,
474
+ "step": 360
475
+ },
476
+ {
477
+ "epoch": 16.0,
478
+ "grad_norm": 0.513087272644043,
479
+ "learning_rate": 6.666666666666667e-06,
480
+ "loss": 0.1023,
481
+ "step": 384
482
+ },
483
+ {
484
+ "epoch": 16.0,
485
+ "eval_acc_macro": 0.8235455386482644,
486
+ "eval_acc_micro": 0.8347425057647321,
487
+ "eval_auc_macro": 0.9885335456453125,
488
+ "eval_auc_micro": 0.9898872283756374,
489
+ "eval_f1_at_5": 0.4275326584009282,
490
+ "eval_f1_at_8": 0.2915919235644548,
491
+ "eval_f1_macro": 0.8999181306237283,
492
+ "eval_f1_micro": 0.9099287808964466,
493
+ "eval_loss": 0.09510227292776108,
494
+ "eval_prec_at_5": 0.27374429223744295,
495
+ "eval_prec_at_8": 0.1713755707762557,
496
+ "eval_prec_macro": 0.9168967179225254,
497
+ "eval_prec_micro": 0.9195596951735038,
498
+ "eval_rec_at_5": 0.9756468797564688,
499
+ "eval_rec_at_8": 0.9767884322678843,
500
+ "eval_rec_macro": 0.8861519346844252,
501
+ "eval_rec_micro": 0.9004975124377362,
502
+ "eval_runtime": 3.3019,
503
+ "eval_samples_per_second": 265.298,
504
+ "eval_steps_per_second": 33.314,
505
+ "step": 384
506
+ },
507
+ {
508
+ "epoch": 17.0,
509
+ "grad_norm": 0.44329148530960083,
510
+ "learning_rate": 6.3888888888888885e-06,
511
+ "loss": 0.0966,
512
+ "step": 408
513
+ },
514
+ {
515
+ "epoch": 17.0,
516
+ "eval_acc_macro": 0.8282861708589082,
517
+ "eval_acc_micro": 0.8414164742108667,
518
+ "eval_auc_macro": 0.9900771631031816,
519
+ "eval_auc_micro": 0.9908838392748628,
520
+ "eval_f1_at_5": 0.4275326584009282,
521
+ "eval_f1_at_8": 0.29182390543479947,
522
+ "eval_f1_macro": 0.903024454767273,
523
+ "eval_f1_micro": 0.9138795986621309,
524
+ "eval_loss": 0.09051786363124847,
525
+ "eval_prec_at_5": 0.27374429223744295,
526
+ "eval_prec_at_8": 0.17151826484018265,
527
+ "eval_prec_macro": 0.9150389365619009,
528
+ "eval_prec_micro": 0.9215851602022831,
529
+ "eval_rec_at_5": 0.9756468797564688,
530
+ "eval_rec_at_8": 0.977359208523592,
531
+ "eval_rec_macro": 0.8934998678358861,
532
+ "eval_rec_micro": 0.9063018242121967,
533
+ "eval_runtime": 3.2033,
534
+ "eval_samples_per_second": 273.471,
535
+ "eval_steps_per_second": 34.34,
536
+ "step": 408
537
+ },
538
+ {
539
+ "epoch": 18.0,
540
+ "grad_norm": 0.5752924084663391,
541
+ "learning_rate": 6.111111111111112e-06,
542
+ "loss": 0.0907,
543
+ "step": 432
544
+ },
545
+ {
546
+ "epoch": 18.0,
547
+ "eval_acc_macro": 0.8284485888604599,
548
+ "eval_acc_micro": 0.8406466512701432,
549
+ "eval_auc_macro": 0.9899844969634316,
550
+ "eval_auc_micro": 0.9914398385606178,
551
+ "eval_f1_at_5": 0.4275326584009282,
552
+ "eval_f1_at_8": 0.29182390543479947,
553
+ "eval_f1_macro": 0.9033529953273461,
554
+ "eval_f1_micro": 0.9134253450438383,
555
+ "eval_loss": 0.08615443110466003,
556
+ "eval_prec_at_5": 0.27374429223744295,
557
+ "eval_prec_at_8": 0.17151826484018265,
558
+ "eval_prec_macro": 0.9148533164182675,
559
+ "eval_prec_micro": 0.9215189873416944,
560
+ "eval_rec_at_5": 0.9756468797564688,
561
+ "eval_rec_at_8": 0.977359208523592,
562
+ "eval_rec_macro": 0.8936470636992689,
563
+ "eval_rec_micro": 0.9054726368158453,
564
+ "eval_runtime": 3.2471,
565
+ "eval_samples_per_second": 269.782,
566
+ "eval_steps_per_second": 33.877,
567
+ "step": 432
568
+ },
569
+ {
570
+ "epoch": 19.0,
571
+ "grad_norm": 0.3273939788341522,
572
+ "learning_rate": 5.833333333333334e-06,
573
+ "loss": 0.0856,
574
+ "step": 456
575
+ },
576
+ {
577
+ "epoch": 19.0,
578
+ "eval_acc_macro": 0.8365219798472802,
579
+ "eval_acc_micro": 0.8460943542149384,
580
+ "eval_auc_macro": 0.990006520319756,
581
+ "eval_auc_micro": 0.991490537854463,
582
+ "eval_f1_at_5": 0.4275326584009282,
583
+ "eval_f1_at_8": 0.29182390543479947,
584
+ "eval_f1_macro": 0.908175633993403,
585
+ "eval_f1_micro": 0.9166317553413559,
586
+ "eval_loss": 0.08281172811985016,
587
+ "eval_prec_at_5": 0.27374429223744295,
588
+ "eval_prec_at_8": 0.17151826484018265,
589
+ "eval_prec_macro": 0.9221522942691044,
590
+ "eval_prec_micro": 0.9263336155799384,
591
+ "eval_rec_at_5": 0.9756468797564688,
592
+ "eval_rec_at_8": 0.977359208523592,
593
+ "eval_rec_macro": 0.8964813104500736,
594
+ "eval_rec_micro": 0.9071310116085483,
595
+ "eval_runtime": 3.2394,
596
+ "eval_samples_per_second": 270.419,
597
+ "eval_steps_per_second": 33.957,
598
+ "step": 456
599
+ },
600
+ {
601
+ "epoch": 20.0,
602
+ "grad_norm": 0.4600836932659149,
603
+ "learning_rate": 5.555555555555557e-06,
604
+ "loss": 0.081,
605
+ "step": 480
606
+ },
607
+ {
608
+ "epoch": 20.0,
609
+ "eval_acc_macro": 0.8341810383521641,
610
+ "eval_acc_micro": 0.8469778117826436,
611
+ "eval_auc_macro": 0.9899728237387516,
612
+ "eval_auc_micro": 0.9915228820436822,
613
+ "eval_f1_at_5": 0.4278659031068435,
614
+ "eval_f1_at_8": 0.2920473967500829,
615
+ "eval_f1_macro": 0.9068497360775286,
616
+ "eval_f1_micro": 0.9171499585749033,
617
+ "eval_loss": 0.07994194328784943,
618
+ "eval_prec_at_5": 0.27397260273972607,
619
+ "eval_prec_at_8": 0.1716609589041096,
620
+ "eval_prec_macro": 0.9087038321630772,
621
+ "eval_prec_micro": 0.9163907284767453,
622
+ "eval_rec_at_5": 0.9762176560121765,
623
+ "eval_rec_at_8": 0.9777397260273972,
624
+ "eval_rec_macro": 0.9059843761789805,
625
+ "eval_rec_micro": 0.9179104477611179,
626
+ "eval_runtime": 3.1998,
627
+ "eval_samples_per_second": 273.764,
628
+ "eval_steps_per_second": 34.377,
629
+ "step": 480
630
+ },
631
+ {
632
+ "epoch": 21.0,
633
+ "grad_norm": 0.5419149398803711,
634
+ "learning_rate": 5.2777777777777785e-06,
635
+ "loss": 0.0773,
636
+ "step": 504
637
+ },
638
+ {
639
+ "epoch": 21.0,
640
+ "eval_acc_macro": 0.8302896897341306,
641
+ "eval_acc_micro": 0.8448947778643144,
642
+ "eval_auc_macro": 0.9902784944925817,
643
+ "eval_auc_micro": 0.9916112716737253,
644
+ "eval_f1_at_5": 0.4275326584009282,
645
+ "eval_f1_at_8": 0.29182390543479947,
646
+ "eval_f1_macro": 0.9041333779773183,
647
+ "eval_f1_micro": 0.9159273341782074,
648
+ "eval_loss": 0.07834368944168091,
649
+ "eval_prec_at_5": 0.27374429223744295,
650
+ "eval_prec_at_8": 0.17151826484018265,
651
+ "eval_prec_macro": 0.9276589497355273,
652
+ "eval_prec_micro": 0.9336778639103416,
653
+ "eval_rec_at_5": 0.9756468797564688,
654
+ "eval_rec_at_8": 0.977359208523592,
655
+ "eval_rec_macro": 0.8839118707511296,
656
+ "eval_rec_micro": 0.8988391376450332,
657
+ "eval_runtime": 3.1453,
658
+ "eval_samples_per_second": 278.513,
659
+ "eval_steps_per_second": 34.973,
660
+ "step": 504
661
+ },
662
+ {
663
+ "epoch": 22.0,
664
+ "grad_norm": 0.3567904829978943,
665
+ "learning_rate": 5e-06,
666
+ "loss": 0.0729,
667
+ "step": 528
668
+ },
669
+ {
670
+ "epoch": 22.0,
671
+ "eval_acc_macro": 0.8348824235565738,
672
+ "eval_acc_micro": 0.8433919022153672,
673
+ "eval_auc_macro": 0.9903521725932115,
674
+ "eval_auc_micro": 0.9919912936214408,
675
+ "eval_f1_at_5": 0.4275326584009282,
676
+ "eval_f1_at_8": 0.2920473967500829,
677
+ "eval_f1_macro": 0.9076004004647569,
678
+ "eval_f1_micro": 0.915043514297479,
679
+ "eval_loss": 0.07567035406827927,
680
+ "eval_prec_at_5": 0.27374429223744295,
681
+ "eval_prec_at_8": 0.1716609589041096,
682
+ "eval_prec_macro": 0.9101730173612985,
683
+ "eval_prec_micro": 0.9146644573321528,
684
+ "eval_rec_at_5": 0.9756468797564688,
685
+ "eval_rec_at_8": 0.9777397260273972,
686
+ "eval_rec_macro": 0.9057605150279991,
687
+ "eval_rec_micro": 0.9154228855720634,
688
+ "eval_runtime": 3.1937,
689
+ "eval_samples_per_second": 274.293,
690
+ "eval_steps_per_second": 34.443,
691
+ "step": 528
692
+ },
693
+ {
694
+ "epoch": 23.0,
695
+ "grad_norm": 0.4837111234664917,
696
+ "learning_rate": 4.722222222222222e-06,
697
+ "loss": 0.0698,
698
+ "step": 552
699
+ },
700
+ {
701
+ "epoch": 23.0,
702
+ "eval_acc_macro": 0.8374765617625478,
703
+ "eval_acc_micro": 0.8441947565542438,
704
+ "eval_auc_macro": 0.9904772253291115,
705
+ "eval_auc_micro": 0.9921461536789155,
706
+ "eval_f1_at_5": 0.4278659031068434,
707
+ "eval_f1_at_8": 0.2920473967500829,
708
+ "eval_f1_macro": 0.9092344767866304,
709
+ "eval_f1_micro": 0.9155158407797794,
710
+ "eval_loss": 0.07319281995296478,
711
+ "eval_prec_at_5": 0.273972602739726,
712
+ "eval_prec_at_8": 0.1716609589041096,
713
+ "eval_prec_macro": 0.8914748226041258,
714
+ "eval_prec_micro": 0.8972929936305017,
715
+ "eval_rec_at_5": 0.9762176560121765,
716
+ "eval_rec_at_8": 0.9777397260273972,
717
+ "eval_rec_macro": 0.9284071354304921,
718
+ "eval_rec_micro": 0.9344941956881481,
719
+ "eval_runtime": 3.1304,
720
+ "eval_samples_per_second": 279.84,
721
+ "eval_steps_per_second": 35.14,
722
+ "step": 552
723
+ },
724
+ {
725
+ "epoch": 24.0,
726
+ "grad_norm": 0.39113664627075195,
727
+ "learning_rate": 4.444444444444444e-06,
728
+ "loss": 0.0666,
729
+ "step": 576
730
+ },
731
+ {
732
+ "epoch": 24.0,
733
+ "eval_acc_macro": 0.8381100355802934,
734
+ "eval_acc_micro": 0.844077961019427,
735
+ "eval_auc_macro": 0.9903966871163578,
736
+ "eval_auc_micro": 0.9918843706818725,
737
+ "eval_f1_at_5": 0.4275326584009282,
738
+ "eval_f1_at_8": 0.2920473967500829,
739
+ "eval_f1_macro": 0.9097231670248208,
740
+ "eval_f1_micro": 0.9154471544714704,
741
+ "eval_loss": 0.07271166890859604,
742
+ "eval_prec_at_5": 0.27374429223744295,
743
+ "eval_prec_at_8": 0.1716609589041096,
744
+ "eval_prec_macro": 0.8946829660987885,
745
+ "eval_prec_micro": 0.8979266347686684,
746
+ "eval_rec_at_5": 0.9756468797564688,
747
+ "eval_rec_at_8": 0.9777397260273972,
748
+ "eval_rec_macro": 0.9261163381279823,
749
+ "eval_rec_micro": 0.9336650082917965,
750
+ "eval_runtime": 3.2473,
751
+ "eval_samples_per_second": 269.762,
752
+ "eval_steps_per_second": 33.874,
753
+ "step": 576
754
+ },
755
+ {
756
+ "epoch": 25.0,
757
+ "grad_norm": 0.623213529586792,
758
+ "learning_rate": 4.166666666666667e-06,
759
+ "loss": 0.0635,
760
+ "step": 600
761
+ },
762
+ {
763
+ "epoch": 25.0,
764
+ "eval_acc_macro": 0.8378381654750737,
765
+ "eval_acc_micro": 0.8448795180722255,
766
+ "eval_auc_macro": 0.9903437445438174,
767
+ "eval_auc_micro": 0.991977126331948,
768
+ "eval_f1_at_5": 0.4275326584009282,
769
+ "eval_f1_at_8": 0.2920473967500829,
770
+ "eval_f1_macro": 0.9095042466910366,
771
+ "eval_f1_micro": 0.915918367346864,
772
+ "eval_loss": 0.07025843113660812,
773
+ "eval_prec_at_5": 0.27374429223744295,
774
+ "eval_prec_at_8": 0.1716609589041096,
775
+ "eval_prec_macro": 0.8980030893359562,
776
+ "eval_prec_micro": 0.9019292604500883,
777
+ "eval_rec_at_5": 0.9756468797564688,
778
+ "eval_rec_at_8": 0.9777397260273972,
779
+ "eval_rec_macro": 0.921910312969234,
780
+ "eval_rec_micro": 0.9303482587063905,
781
+ "eval_runtime": 3.2584,
782
+ "eval_samples_per_second": 268.843,
783
+ "eval_steps_per_second": 33.759,
784
+ "step": 600
785
+ },
786
+ {
787
+ "epoch": 26.0,
788
+ "grad_norm": 0.5504964590072632,
789
+ "learning_rate": 3.88888888888889e-06,
790
+ "loss": 0.0608,
791
+ "step": 624
792
+ },
793
+ {
794
+ "epoch": 26.0,
795
+ "eval_acc_macro": 0.8420968191195207,
796
+ "eval_acc_micro": 0.8508705526115934,
797
+ "eval_auc_macro": 0.9903203584733035,
798
+ "eval_auc_micro": 0.9918962213076746,
799
+ "eval_f1_at_5": 0.4271994046239887,
800
+ "eval_f1_at_8": 0.2920473967500829,
801
+ "eval_f1_macro": 0.911966406429931,
802
+ "eval_f1_micro": 0.9194274028629105,
803
+ "eval_loss": 0.06913089752197266,
804
+ "eval_prec_at_5": 0.27351598173515984,
805
+ "eval_prec_at_8": 0.1716609589041096,
806
+ "eval_prec_macro": 0.8987539781949079,
807
+ "eval_prec_micro": 0.9071832122678848,
808
+ "eval_rec_at_5": 0.975076103500761,
809
+ "eval_rec_at_8": 0.9777397260273972,
810
+ "eval_rec_macro": 0.9260958603602073,
811
+ "eval_rec_micro": 0.9320066334990935,
812
+ "eval_runtime": 3.2691,
813
+ "eval_samples_per_second": 267.962,
814
+ "eval_steps_per_second": 33.648,
815
+ "step": 624
816
+ },
817
+ {
818
+ "epoch": 27.0,
819
+ "grad_norm": 0.4138895571231842,
820
+ "learning_rate": 3.6111111111111115e-06,
821
+ "loss": 0.059,
822
+ "step": 648
823
+ },
824
+ {
825
+ "epoch": 27.0,
826
+ "eval_acc_macro": 0.8343605198418258,
827
+ "eval_acc_micro": 0.8466819221967318,
828
+ "eval_auc_macro": 0.9905189753061885,
829
+ "eval_auc_micro": 0.9920985729708075,
830
+ "eval_f1_at_5": 0.4278659031068434,
831
+ "eval_f1_at_8": 0.2920473967500829,
832
+ "eval_f1_macro": 0.9071167088535305,
833
+ "eval_f1_micro": 0.9169764560098375,
834
+ "eval_loss": 0.068180613219738,
835
+ "eval_prec_at_5": 0.273972602739726,
836
+ "eval_prec_at_8": 0.1716609589041096,
837
+ "eval_prec_macro": 0.905623991446007,
838
+ "eval_prec_micro": 0.913580246913505,
839
+ "eval_rec_at_5": 0.9762176560121765,
840
+ "eval_rec_at_8": 0.9777397260273972,
841
+ "eval_rec_macro": 0.9092956538553008,
842
+ "eval_rec_micro": 0.9203980099501724,
843
+ "eval_runtime": 3.2333,
844
+ "eval_samples_per_second": 270.935,
845
+ "eval_steps_per_second": 34.021,
846
+ "step": 648
847
+ },
848
+ {
849
+ "epoch": 28.0,
850
+ "grad_norm": 0.4782184958457947,
851
+ "learning_rate": 3.3333333333333333e-06,
852
+ "loss": 0.0558,
853
+ "step": 672
854
+ },
855
+ {
856
+ "epoch": 28.0,
857
+ "eval_acc_macro": 0.8388794744893889,
858
+ "eval_acc_micro": 0.8464419475654796,
859
+ "eval_auc_macro": 0.9907869944842939,
860
+ "eval_auc_micro": 0.9923112605180986,
861
+ "eval_f1_at_5": 0.4281991387591175,
862
+ "eval_f1_at_8": 0.2920473967500829,
863
+ "eval_f1_macro": 0.9097311444189383,
864
+ "eval_f1_micro": 0.9168356997970858,
865
+ "eval_loss": 0.06692954152822495,
866
+ "eval_prec_at_5": 0.2742009132420091,
867
+ "eval_prec_at_8": 0.1716609589041096,
868
+ "eval_prec_macro": 0.8914388177432641,
869
+ "eval_prec_micro": 0.8975377283557666,
870
+ "eval_rec_at_5": 0.9767884322678843,
871
+ "eval_rec_at_8": 0.9777397260273972,
872
+ "eval_rec_macro": 0.9295486988792282,
873
+ "eval_rec_micro": 0.9369817578772025,
874
+ "eval_runtime": 3.1471,
875
+ "eval_samples_per_second": 278.355,
876
+ "eval_steps_per_second": 34.953,
877
+ "step": 672
878
+ },
879
+ {
880
+ "epoch": 29.0,
881
+ "grad_norm": 0.4504872262477875,
882
+ "learning_rate": 3.055555555555556e-06,
883
+ "loss": 0.0538,
884
+ "step": 696
885
+ },
886
+ {
887
+ "epoch": 29.0,
888
+ "eval_acc_macro": 0.838778704972707,
889
+ "eval_acc_micro": 0.849157733537454,
890
+ "eval_auc_macro": 0.9902459134171048,
891
+ "eval_auc_micro": 0.9921254819105988,
892
+ "eval_f1_at_5": 0.4278476261962846,
893
+ "eval_f1_at_8": 0.2920473967500829,
894
+ "eval_f1_macro": 0.909828962756635,
895
+ "eval_f1_micro": 0.9184265010351206,
896
+ "eval_loss": 0.06644343584775925,
897
+ "eval_prec_at_5": 0.27397260273972607,
898
+ "eval_prec_at_8": 0.1716609589041096,
899
+ "eval_prec_macro": 0.9086581525308847,
900
+ "eval_prec_micro": 0.9172870140611317,
901
+ "eval_rec_at_5": 0.976027397260274,
902
+ "eval_rec_at_8": 0.9777397260273972,
903
+ "eval_rec_macro": 0.9113050956999702,
904
+ "eval_rec_micro": 0.9195688225538209,
905
+ "eval_runtime": 3.1423,
906
+ "eval_samples_per_second": 278.777,
907
+ "eval_steps_per_second": 35.006,
908
+ "step": 696
909
+ },
910
+ {
911
+ "epoch": 30.0,
912
+ "grad_norm": 0.4144255220890045,
913
+ "learning_rate": 2.7777777777777783e-06,
914
+ "loss": 0.0514,
915
+ "step": 720
916
+ },
917
+ {
918
+ "epoch": 30.0,
919
+ "eval_acc_macro": 0.8414238560509153,
920
+ "eval_acc_micro": 0.8515981735159169,
921
+ "eval_auc_macro": 0.9904232681518291,
922
+ "eval_auc_micro": 0.9915424845826032,
923
+ "eval_f1_at_5": 0.4278659031068435,
924
+ "eval_f1_at_8": 0.29182390543479947,
925
+ "eval_f1_macro": 0.9115867876146071,
926
+ "eval_f1_micro": 0.9198520345252017,
927
+ "eval_loss": 0.06602790951728821,
928
+ "eval_prec_at_5": 0.27397260273972607,
929
+ "eval_prec_at_8": 0.17151826484018265,
930
+ "eval_prec_macro": 0.9028158103948029,
931
+ "eval_prec_micro": 0.9119804400977252,
932
+ "eval_rec_at_5": 0.9762176560121765,
933
+ "eval_rec_at_8": 0.977359208523592,
934
+ "eval_rec_macro": 0.9208453689121621,
935
+ "eval_rec_micro": 0.9278606965173359,
936
+ "eval_runtime": 3.2592,
937
+ "eval_samples_per_second": 268.778,
938
+ "eval_steps_per_second": 33.751,
939
+ "step": 720
940
+ },
941
+ {
942
+ "epoch": 31.0,
943
+ "grad_norm": 0.5076408386230469,
944
+ "learning_rate": 2.5e-06,
945
+ "loss": 0.0486,
946
+ "step": 744
947
+ },
948
+ {
949
+ "epoch": 31.0,
950
+ "eval_acc_macro": 0.8417559403467925,
951
+ "eval_acc_micro": 0.8513097072418451,
952
+ "eval_auc_macro": 0.9902567259413712,
953
+ "eval_auc_micro": 0.9916356857449267,
954
+ "eval_f1_at_5": 0.42747783226694747,
955
+ "eval_f1_at_8": 0.2918154230125642,
956
+ "eval_f1_macro": 0.9114967651862192,
957
+ "eval_f1_micro": 0.9196837286724161,
958
+ "eval_loss": 0.06532150506973267,
959
+ "eval_prec_at_5": 0.27374429223744295,
960
+ "eval_prec_at_8": 0.17151826484018265,
961
+ "eval_prec_macro": 0.9135306848535077,
962
+ "eval_prec_micro": 0.9231411862990039,
963
+ "eval_rec_at_5": 0.975076103500761,
964
+ "eval_rec_at_8": 0.9771689497716894,
965
+ "eval_rec_macro": 0.9101782005755276,
966
+ "eval_rec_micro": 0.9162520729684148,
967
+ "eval_runtime": 3.154,
968
+ "eval_samples_per_second": 277.742,
969
+ "eval_steps_per_second": 34.876,
970
+ "step": 744
971
+ },
972
+ {
973
+ "epoch": 32.0,
974
+ "grad_norm": 0.39073243737220764,
975
+ "learning_rate": 2.222222222222222e-06,
976
+ "loss": 0.0473,
977
+ "step": 768
978
+ },
979
+ {
980
+ "epoch": 32.0,
981
+ "eval_acc_macro": 0.8365248835222155,
982
+ "eval_acc_micro": 0.8455034588777213,
983
+ "eval_auc_macro": 0.9902217568533902,
984
+ "eval_auc_micro": 0.991443937273301,
985
+ "eval_f1_at_5": 0.42747783226694747,
986
+ "eval_f1_at_8": 0.2920473967500829,
987
+ "eval_f1_macro": 0.9082513467521097,
988
+ "eval_f1_micro": 0.9162848812993822,
989
+ "eval_loss": 0.06487824022769928,
990
+ "eval_prec_at_5": 0.27374429223744295,
991
+ "eval_prec_at_8": 0.1716609589041096,
992
+ "eval_prec_macro": 0.9133935694619835,
993
+ "eval_prec_micro": 0.9205020920501321,
994
+ "eval_rec_at_5": 0.975076103500761,
995
+ "eval_rec_at_8": 0.9777397260273972,
996
+ "eval_rec_macro": 0.904152383416957,
997
+ "eval_rec_micro": 0.9121061359866574,
998
+ "eval_runtime": 3.241,
999
+ "eval_samples_per_second": 270.288,
1000
+ "eval_steps_per_second": 33.94,
1001
+ "step": 768
1002
+ },
1003
+ {
1004
+ "epoch": 33.0,
1005
+ "grad_norm": 0.4108600318431854,
1006
+ "learning_rate": 1.944444444444445e-06,
1007
+ "loss": 0.0452,
1008
+ "step": 792
1009
+ },
1010
+ {
1011
+ "epoch": 33.0,
1012
+ "eval_acc_macro": 0.8342283813769025,
1013
+ "eval_acc_micro": 0.8460365853657892,
1014
+ "eval_auc_macro": 0.9900494801767596,
1015
+ "eval_auc_micro": 0.9917285304974518,
1016
+ "eval_f1_at_5": 0.4278476261962846,
1017
+ "eval_f1_at_8": 0.2920473967500829,
1018
+ "eval_f1_macro": 0.9069117318096943,
1019
+ "eval_f1_micro": 0.9165978530139621,
1020
+ "eval_loss": 0.06424969434738159,
1021
+ "eval_prec_at_5": 0.27397260273972607,
1022
+ "eval_prec_at_8": 0.1716609589041096,
1023
+ "eval_prec_macro": 0.9043399542883956,
1024
+ "eval_prec_micro": 0.912828947368346,
1025
+ "eval_rec_at_5": 0.976027397260274,
1026
+ "eval_rec_at_8": 0.9777397260273972,
1027
+ "eval_rec_macro": 0.9102122008404995,
1028
+ "eval_rec_micro": 0.9203980099501724,
1029
+ "eval_runtime": 3.1728,
1030
+ "eval_samples_per_second": 276.1,
1031
+ "eval_steps_per_second": 34.67,
1032
+ "step": 792
1033
+ },
1034
+ {
1035
+ "epoch": 34.0,
1036
+ "grad_norm": 0.6398113369941711,
1037
+ "learning_rate": 1.6666666666666667e-06,
1038
+ "loss": 0.0441,
1039
+ "step": 816
1040
+ },
1041
+ {
1042
+ "epoch": 34.0,
1043
+ "eval_acc_macro": 0.8402010441768392,
1044
+ "eval_acc_micro": 0.8504176157934054,
1045
+ "eval_auc_macro": 0.9899877083743055,
1046
+ "eval_auc_micro": 0.9911509684188842,
1047
+ "eval_f1_at_5": 0.42747783226694747,
1048
+ "eval_f1_at_8": 0.29182390543479947,
1049
+ "eval_f1_macro": 0.9105990068832329,
1050
+ "eval_f1_micro": 0.9191629052112499,
1051
+ "eval_loss": 0.06337956339120865,
1052
+ "eval_prec_at_5": 0.27374429223744295,
1053
+ "eval_prec_at_8": 0.17151826484018265,
1054
+ "eval_prec_macro": 0.9012107317710064,
1055
+ "eval_prec_micro": 0.9098294069861161,
1056
+ "eval_rec_at_5": 0.975076103500761,
1057
+ "eval_rec_at_8": 0.977359208523592,
1058
+ "eval_rec_macro": 0.921097280138882,
1059
+ "eval_rec_micro": 0.9286898839136875,
1060
+ "eval_runtime": 3.2382,
1061
+ "eval_samples_per_second": 270.523,
1062
+ "eval_steps_per_second": 33.97,
1063
+ "step": 816
1064
+ },
1065
+ {
1066
+ "epoch": 35.0,
1067
+ "grad_norm": 0.5846818089485168,
1068
+ "learning_rate": 1.3888888888888892e-06,
1069
+ "loss": 0.0423,
1070
+ "step": 840
1071
+ },
1072
+ {
1073
+ "epoch": 35.0,
1074
+ "eval_acc_macro": 0.8392593134167763,
1075
+ "eval_acc_micro": 0.8493788819875117,
1076
+ "eval_auc_macro": 0.9896914288641514,
1077
+ "eval_auc_micro": 0.9914127959671519,
1078
+ "eval_f1_at_5": 0.42747783226694747,
1079
+ "eval_f1_at_8": 0.2920473967500829,
1080
+ "eval_f1_macro": 0.9100464733016533,
1081
+ "eval_f1_micro": 0.9185558354323325,
1082
+ "eval_loss": 0.06289488822221756,
1083
+ "eval_prec_at_5": 0.27374429223744295,
1084
+ "eval_prec_at_8": 0.1716609589041096,
1085
+ "eval_prec_macro": 0.9212797353863423,
1086
+ "eval_prec_micro": 0.9302721088434582,
1087
+ "eval_rec_at_5": 0.975076103500761,
1088
+ "eval_rec_at_8": 0.9777397260273972,
1089
+ "eval_rec_macro": 0.9000358120655063,
1090
+ "eval_rec_micro": 0.9071310116085483,
1091
+ "eval_runtime": 3.1859,
1092
+ "eval_samples_per_second": 274.961,
1093
+ "eval_steps_per_second": 34.527,
1094
+ "step": 840
1095
+ },
1096
+ {
1097
+ "epoch": 36.0,
1098
+ "grad_norm": 0.47924453020095825,
1099
+ "learning_rate": 1.111111111111111e-06,
1100
+ "loss": 0.0408,
1101
+ "step": 864
1102
+ },
1103
+ {
1104
+ "epoch": 36.0,
1105
+ "eval_acc_macro": 0.839620505353177,
1106
+ "eval_acc_micro": 0.8521072796934213,
1107
+ "eval_auc_macro": 0.9895009786225338,
1108
+ "eval_auc_micro": 0.990984346838057,
1109
+ "eval_f1_at_5": 0.42747783226694747,
1110
+ "eval_f1_at_8": 0.2920473967500829,
1111
+ "eval_f1_macro": 0.9100739843959514,
1112
+ "eval_f1_micro": 0.9201489449730311,
1113
+ "eval_loss": 0.06270425021648407,
1114
+ "eval_prec_at_5": 0.27374429223744295,
1115
+ "eval_prec_at_8": 0.1716609589041096,
1116
+ "eval_prec_macro": 0.9077928816879504,
1117
+ "eval_prec_micro": 0.9182493806770505,
1118
+ "eval_rec_at_5": 0.975076103500761,
1119
+ "eval_rec_at_8": 0.9777397260273972,
1120
+ "eval_rec_macro": 0.9134800253293897,
1121
+ "eval_rec_micro": 0.9220563847428754,
1122
+ "eval_runtime": 3.2485,
1123
+ "eval_samples_per_second": 269.663,
1124
+ "eval_steps_per_second": 33.862,
1125
+ "step": 864
1126
+ },
1127
+ {
1128
+ "epoch": 37.0,
1129
+ "grad_norm": 0.45999017357826233,
1130
+ "learning_rate": 8.333333333333333e-07,
1131
+ "loss": 0.0389,
1132
+ "step": 888
1133
+ },
1134
+ {
1135
+ "epoch": 37.0,
1136
+ "eval_acc_macro": 0.838151465593231,
1137
+ "eval_acc_micro": 0.8484848484847842,
1138
+ "eval_auc_macro": 0.9893868618238587,
1139
+ "eval_auc_micro": 0.9910452038111611,
1140
+ "eval_f1_at_5": 0.42714459980714947,
1141
+ "eval_f1_at_8": 0.2920473967500829,
1142
+ "eval_f1_macro": 0.9091468843830479,
1143
+ "eval_f1_micro": 0.9180327868851706,
1144
+ "eval_loss": 0.0631389170885086,
1145
+ "eval_prec_at_5": 0.27351598173515984,
1146
+ "eval_prec_at_8": 0.1716609589041096,
1147
+ "eval_prec_macro": 0.9002505510433269,
1148
+ "eval_prec_micro": 0.9076175040517903,
1149
+ "eval_rec_at_5": 0.9745053272450532,
1150
+ "eval_rec_at_8": 0.9777397260273972,
1151
+ "eval_rec_macro": 0.9191609975742806,
1152
+ "eval_rec_micro": 0.9286898839136875,
1153
+ "eval_runtime": 3.1965,
1154
+ "eval_samples_per_second": 274.047,
1155
+ "eval_steps_per_second": 34.412,
1156
+ "step": 888
1157
+ },
1158
+ {
1159
+ "epoch": 38.0,
1160
+ "grad_norm": 0.5668067336082458,
1161
+ "learning_rate": 5.555555555555555e-07,
1162
+ "loss": 0.038,
1163
+ "step": 912
1164
+ },
1165
+ {
1166
+ "epoch": 38.0,
1167
+ "eval_acc_macro": 0.8442263088885594,
1168
+ "eval_acc_micro": 0.8538985616956204,
1169
+ "eval_auc_macro": 0.989457040438969,
1170
+ "eval_auc_micro": 0.9912924631089128,
1171
+ "eval_f1_at_5": 0.42812596936069724,
1172
+ "eval_f1_at_8": 0.2920473967500829,
1173
+ "eval_f1_macro": 0.9129580069246931,
1174
+ "eval_f1_micro": 0.9211923233972297,
1175
+ "eval_loss": 0.062041375786066055,
1176
+ "eval_prec_at_5": 0.2742009132420091,
1177
+ "eval_prec_at_8": 0.1716609589041096,
1178
+ "eval_prec_macro": 0.9013584195385889,
1179
+ "eval_prec_micro": 0.907481898632268,
1180
+ "eval_rec_at_5": 0.976027397260274,
1181
+ "eval_rec_at_8": 0.9777397260273972,
1182
+ "eval_rec_macro": 0.9261603815096769,
1183
+ "eval_rec_micro": 0.9353233830844995,
1184
+ "eval_runtime": 3.186,
1185
+ "eval_samples_per_second": 274.957,
1186
+ "eval_steps_per_second": 34.527,
1187
+ "step": 912
1188
+ }
1189
+ ],
1190
+ "logging_steps": 500,
1191
+ "max_steps": 960,
1192
+ "num_input_tokens_seen": 0,
1193
+ "num_train_epochs": 40,
1194
+ "save_steps": 500,
1195
+ "total_flos": 1.533864671059968e+16,
1196
+ "train_batch_size": 32,
1197
+ "trial_name": null,
1198
+ "trial_params": null
1199
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b8e42fc2e78b17009878c149a5babcb96b8477fac0a749918ba2a60e4ace058d
3
+ size 5048
vocab.json ADDED
The diff for this file is too large to render. See raw diff