commit files to HF hub
Browse files- added_tokens.json +18 -0
- config.json +44 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +25 -0
- tokenizer.json +0 -0
- tokenizer_config.json +14 -0
- trainer_state.json +442 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
added_tokens.json
ADDED
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"[ambiance-Negative]": 30526,
|
3 |
+
"[ambiance-Positive]": 30527,
|
4 |
+
"[ambiance-no majority]": 30529,
|
5 |
+
"[ambiance-unknown]": 30528,
|
6 |
+
"[food-Negative]": 30522,
|
7 |
+
"[food-Positive]": 30523,
|
8 |
+
"[food-no majority]": 30525,
|
9 |
+
"[food-unknown]": 30524,
|
10 |
+
"[noise-Negative]": 30530,
|
11 |
+
"[noise-Positive]": 30531,
|
12 |
+
"[noise-no majority]": 30533,
|
13 |
+
"[noise-unknown]": 30532,
|
14 |
+
"[service-Negative]": 30534,
|
15 |
+
"[service-Positive]": 30535,
|
16 |
+
"[service-no majority]": 30537,
|
17 |
+
"[service-unknown]": 30536
|
18 |
+
}
|
config.json
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "CEBaB/lstm.CEBaB.sa.5-class.exclusive.seed_66",
|
3 |
+
"architectures": [
|
4 |
+
"LSTMForFactualCounterfactualSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"bidirectional": true,
|
8 |
+
"classifier_dropout": null,
|
9 |
+
"finetuning_task": "opentable",
|
10 |
+
"gradient_checkpointing": false,
|
11 |
+
"hidden_act": "gelu",
|
12 |
+
"hidden_dropout_prob": 0.1,
|
13 |
+
"hidden_size": 300,
|
14 |
+
"id2label": {
|
15 |
+
"0": 0,
|
16 |
+
"1": 1,
|
17 |
+
"2": 2,
|
18 |
+
"3": 3,
|
19 |
+
"4": 4
|
20 |
+
},
|
21 |
+
"initializer_range": 0.02,
|
22 |
+
"intermediate_size": 3072,
|
23 |
+
"label2id": {
|
24 |
+
"0": 0,
|
25 |
+
"1": 1,
|
26 |
+
"2": 2,
|
27 |
+
"3": 3,
|
28 |
+
"4": 4
|
29 |
+
},
|
30 |
+
"layer_norm_eps": 1e-12,
|
31 |
+
"max_position_embeddings": 512,
|
32 |
+
"model_type": "bert",
|
33 |
+
"num_attention_heads": 12,
|
34 |
+
"num_hidden_layers": 1,
|
35 |
+
"pad_token_id": 0,
|
36 |
+
"position_embedding_type": "absolute",
|
37 |
+
"problem_type": "single_label_classification",
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.21.1",
|
40 |
+
"type_vocab_size": 2,
|
41 |
+
"update_embeddings": false,
|
42 |
+
"use_cache": true,
|
43 |
+
"vocab_size": 30538
|
44 |
+
}
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bb0be9122f69984ae29f040bdbcaaef66a97800c1a4c96d4db7c2905aee107c3
|
3 |
+
size 85600319
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:870ca4d32f4267f00cdfe00163cc3def5f5b8c93f85387e4843dcacee8631369
|
3 |
+
size 42799239
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9e649cd1112166ee83afdc39b48111fa8509d41e461df413319fc78dd3ebdf51
|
3 |
+
size 14503
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:26fee0889bb39f7761726cf7bb1f8cbb7d5aa33a3e330be32cb2ed278fe2d262
|
3 |
+
size 623
|
special_tokens_map.json
ADDED
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"[food-Negative]",
|
4 |
+
"[food-Positive]",
|
5 |
+
"[food-unknown]",
|
6 |
+
"[food-no majority]",
|
7 |
+
"[ambiance-Negative]",
|
8 |
+
"[ambiance-Positive]",
|
9 |
+
"[ambiance-unknown]",
|
10 |
+
"[ambiance-no majority]",
|
11 |
+
"[noise-Negative]",
|
12 |
+
"[noise-Positive]",
|
13 |
+
"[noise-unknown]",
|
14 |
+
"[noise-no majority]",
|
15 |
+
"[service-Negative]",
|
16 |
+
"[service-Positive]",
|
17 |
+
"[service-unknown]",
|
18 |
+
"[service-no majority]"
|
19 |
+
],
|
20 |
+
"cls_token": "[CLS]",
|
21 |
+
"mask_token": "[MASK]",
|
22 |
+
"pad_token": "[PAD]",
|
23 |
+
"sep_token": "[SEP]",
|
24 |
+
"unk_token": "[UNK]"
|
25 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "[CLS]",
|
3 |
+
"do_lower_case": true,
|
4 |
+
"mask_token": "[MASK]",
|
5 |
+
"model_max_length": 512,
|
6 |
+
"name_or_path": "bert-base-uncased",
|
7 |
+
"pad_token": "[PAD]",
|
8 |
+
"sep_token": "[SEP]",
|
9 |
+
"special_tokens_map_file": null,
|
10 |
+
"strip_accents": null,
|
11 |
+
"tokenize_chinese_chars": true,
|
12 |
+
"tokenizer_class": "BertTokenizer",
|
13 |
+
"unk_token": "[UNK]"
|
14 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,442 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.49332907795906067,
|
3 |
+
"best_model_checkpoint": "model_output/e2e_opentable_5_way__approximate__0-shot__seed-66__lstm/checkpoint-650",
|
4 |
+
"epoch": 1.1149228130360207,
|
5 |
+
"global_step": 650,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.03,
|
12 |
+
"learning_rate": 4.9783362218370885e-05,
|
13 |
+
"loss": 43.289,
|
14 |
+
"step": 20
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.07,
|
18 |
+
"learning_rate": 4.956672443674177e-05,
|
19 |
+
"loss": 39.7742,
|
20 |
+
"step": 40
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.09,
|
24 |
+
"eval_distillation_accuracy_counterfactual": 0.3876346844535659,
|
25 |
+
"eval_distillation_accuracy_factual": 0.7116469984607491,
|
26 |
+
"eval_distillation_f1_counterfactual": 0.36789273743176754,
|
27 |
+
"eval_distillation_f1_factual": 0.706355718089049,
|
28 |
+
"eval_groundtruth_accuracy_counterfactual": 0.3170856849666496,
|
29 |
+
"eval_groundtruth_f1_counterfactual": 0.30061734733599244,
|
30 |
+
"eval_groundtruth_f1_factual": 0.5309938705776336,
|
31 |
+
"eval_icace_cosine": 0.7443665266036987,
|
32 |
+
"eval_icace_l2": 0.8025880455970764,
|
33 |
+
"eval_icace_normdiff": 0.6179075837135315,
|
34 |
+
"eval_loss": 21.722654342651367,
|
35 |
+
"eval_runtime": 4.6356,
|
36 |
+
"eval_samples_per_second": 840.887,
|
37 |
+
"eval_steps_per_second": 6.687,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.1,
|
42 |
+
"learning_rate": 4.935008665511265e-05,
|
43 |
+
"loss": 36.1211,
|
44 |
+
"step": 60
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.14,
|
48 |
+
"learning_rate": 4.913344887348354e-05,
|
49 |
+
"loss": 33.0051,
|
50 |
+
"step": 80
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.17,
|
54 |
+
"learning_rate": 4.891681109185442e-05,
|
55 |
+
"loss": 29.0866,
|
56 |
+
"step": 100
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.17,
|
60 |
+
"eval_distillation_accuracy_counterfactual": 0.4179066187788609,
|
61 |
+
"eval_distillation_accuracy_factual": 0.7213955874807594,
|
62 |
+
"eval_distillation_f1_counterfactual": 0.4011779697256409,
|
63 |
+
"eval_distillation_f1_factual": 0.7130234220148569,
|
64 |
+
"eval_groundtruth_accuracy_counterfactual": 0.36198050282196,
|
65 |
+
"eval_groundtruth_f1_counterfactual": 0.34119540585580543,
|
66 |
+
"eval_groundtruth_f1_factual": 0.533506870286257,
|
67 |
+
"eval_icace_cosine": 0.6045525074005127,
|
68 |
+
"eval_icace_l2": 0.7391046285629272,
|
69 |
+
"eval_icace_normdiff": 0.5667964220046997,
|
70 |
+
"eval_loss": 15.981736183166504,
|
71 |
+
"eval_runtime": 4.6279,
|
72 |
+
"eval_samples_per_second": 842.278,
|
73 |
+
"eval_steps_per_second": 6.698,
|
74 |
+
"step": 100
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.21,
|
78 |
+
"learning_rate": 4.8700173310225307e-05,
|
79 |
+
"loss": 24.2452,
|
80 |
+
"step": 120
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.24,
|
84 |
+
"learning_rate": 4.848353552859619e-05,
|
85 |
+
"loss": 20.9734,
|
86 |
+
"step": 140
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.26,
|
90 |
+
"eval_distillation_accuracy_counterfactual": 0.4415084658799384,
|
91 |
+
"eval_distillation_accuracy_factual": 0.6688045151359672,
|
92 |
+
"eval_distillation_f1_counterfactual": 0.39743600677420876,
|
93 |
+
"eval_distillation_f1_factual": 0.6386265115556637,
|
94 |
+
"eval_groundtruth_accuracy_counterfactual": 0.42432016418676244,
|
95 |
+
"eval_groundtruth_f1_counterfactual": 0.37094685868809796,
|
96 |
+
"eval_groundtruth_f1_factual": 0.49392886445649786,
|
97 |
+
"eval_icace_cosine": 0.5455384254455566,
|
98 |
+
"eval_icace_l2": 0.6967277526855469,
|
99 |
+
"eval_icace_normdiff": 0.46749597787857056,
|
100 |
+
"eval_loss": 12.647016525268555,
|
101 |
+
"eval_runtime": 4.5013,
|
102 |
+
"eval_samples_per_second": 865.969,
|
103 |
+
"eval_steps_per_second": 6.887,
|
104 |
+
"step": 150
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.27,
|
108 |
+
"learning_rate": 4.826689774696707e-05,
|
109 |
+
"loss": 19.1703,
|
110 |
+
"step": 160
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.31,
|
114 |
+
"learning_rate": 4.8050259965337955e-05,
|
115 |
+
"loss": 17.6579,
|
116 |
+
"step": 180
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.34,
|
120 |
+
"learning_rate": 4.7833622183708845e-05,
|
121 |
+
"loss": 17.1812,
|
122 |
+
"step": 200
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.34,
|
126 |
+
"eval_distillation_accuracy_counterfactual": 0.46587993842996406,
|
127 |
+
"eval_distillation_accuracy_factual": 0.7129297075423294,
|
128 |
+
"eval_distillation_f1_counterfactual": 0.42585340396044125,
|
129 |
+
"eval_distillation_f1_factual": 0.6801681890460928,
|
130 |
+
"eval_groundtruth_accuracy_counterfactual": 0.44766546947152386,
|
131 |
+
"eval_groundtruth_f1_counterfactual": 0.4011001985251402,
|
132 |
+
"eval_groundtruth_f1_factual": 0.5200271047795778,
|
133 |
+
"eval_icace_cosine": 0.5150377154350281,
|
134 |
+
"eval_icace_l2": 0.6690810322761536,
|
135 |
+
"eval_icace_normdiff": 0.4451918601989746,
|
136 |
+
"eval_loss": 11.362848281860352,
|
137 |
+
"eval_runtime": 4.5371,
|
138 |
+
"eval_samples_per_second": 859.147,
|
139 |
+
"eval_steps_per_second": 6.833,
|
140 |
+
"step": 200
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.38,
|
144 |
+
"learning_rate": 4.761698440207972e-05,
|
145 |
+
"loss": 16.2793,
|
146 |
+
"step": 220
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.41,
|
150 |
+
"learning_rate": 4.740034662045061e-05,
|
151 |
+
"loss": 15.8969,
|
152 |
+
"step": 240
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.43,
|
156 |
+
"eval_distillation_accuracy_counterfactual": 0.48255515649050795,
|
157 |
+
"eval_distillation_accuracy_factual": 0.7508978963571062,
|
158 |
+
"eval_distillation_f1_counterfactual": 0.4392245827288441,
|
159 |
+
"eval_distillation_f1_factual": 0.7287650814209791,
|
160 |
+
"eval_groundtruth_accuracy_counterfactual": 0.4574140584915341,
|
161 |
+
"eval_groundtruth_f1_counterfactual": 0.4058681312407287,
|
162 |
+
"eval_groundtruth_f1_factual": 0.5340988634733079,
|
163 |
+
"eval_icace_cosine": 0.5094192028045654,
|
164 |
+
"eval_icace_l2": 0.6640841960906982,
|
165 |
+
"eval_icace_normdiff": 0.43985918164253235,
|
166 |
+
"eval_loss": 11.062640190124512,
|
167 |
+
"eval_runtime": 4.5851,
|
168 |
+
"eval_samples_per_second": 850.14,
|
169 |
+
"eval_steps_per_second": 6.761,
|
170 |
+
"step": 250
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.45,
|
174 |
+
"learning_rate": 4.7183708838821494e-05,
|
175 |
+
"loss": 15.567,
|
176 |
+
"step": 260
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.48,
|
180 |
+
"learning_rate": 4.6967071057192376e-05,
|
181 |
+
"loss": 15.5899,
|
182 |
+
"step": 280
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.51,
|
186 |
+
"learning_rate": 4.675043327556326e-05,
|
187 |
+
"loss": 15.4491,
|
188 |
+
"step": 300
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.51,
|
192 |
+
"eval_distillation_accuracy_counterfactual": 0.49589533093894306,
|
193 |
+
"eval_distillation_accuracy_factual": 0.7701385325808107,
|
194 |
+
"eval_distillation_f1_counterfactual": 0.4490615827265524,
|
195 |
+
"eval_distillation_f1_factual": 0.7488952781611599,
|
196 |
+
"eval_groundtruth_accuracy_counterfactual": 0.47614161108260644,
|
197 |
+
"eval_groundtruth_f1_counterfactual": 0.4213178039314829,
|
198 |
+
"eval_groundtruth_f1_factual": 0.5320832145764172,
|
199 |
+
"eval_icace_cosine": 0.5026495456695557,
|
200 |
+
"eval_icace_l2": 0.6473169922828674,
|
201 |
+
"eval_icace_normdiff": 0.4321483075618744,
|
202 |
+
"eval_loss": 10.44105339050293,
|
203 |
+
"eval_runtime": 4.6207,
|
204 |
+
"eval_samples_per_second": 843.602,
|
205 |
+
"eval_steps_per_second": 6.709,
|
206 |
+
"step": 300
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.55,
|
210 |
+
"learning_rate": 4.653379549393415e-05,
|
211 |
+
"loss": 15.2629,
|
212 |
+
"step": 320
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.58,
|
216 |
+
"learning_rate": 4.6317157712305025e-05,
|
217 |
+
"loss": 14.8008,
|
218 |
+
"step": 340
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 0.6,
|
222 |
+
"eval_distillation_accuracy_counterfactual": 0.50846587993843,
|
223 |
+
"eval_distillation_accuracy_factual": 0.7747562852744997,
|
224 |
+
"eval_distillation_f1_counterfactual": 0.45682754564452965,
|
225 |
+
"eval_distillation_f1_factual": 0.7559415583306467,
|
226 |
+
"eval_groundtruth_accuracy_counterfactual": 0.49461262185736277,
|
227 |
+
"eval_groundtruth_f1_counterfactual": 0.4326301439786258,
|
228 |
+
"eval_groundtruth_f1_factual": 0.5477637349101722,
|
229 |
+
"eval_icace_cosine": 0.5030019283294678,
|
230 |
+
"eval_icace_l2": 0.6434506177902222,
|
231 |
+
"eval_icace_normdiff": 0.4238852560520172,
|
232 |
+
"eval_loss": 10.315159797668457,
|
233 |
+
"eval_runtime": 4.5363,
|
234 |
+
"eval_samples_per_second": 859.299,
|
235 |
+
"eval_steps_per_second": 6.834,
|
236 |
+
"step": 350
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.62,
|
240 |
+
"learning_rate": 4.6100519930675915e-05,
|
241 |
+
"loss": 14.6876,
|
242 |
+
"step": 360
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 0.65,
|
246 |
+
"learning_rate": 4.58838821490468e-05,
|
247 |
+
"loss": 14.8699,
|
248 |
+
"step": 380
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.69,
|
252 |
+
"learning_rate": 4.566724436741768e-05,
|
253 |
+
"loss": 14.3057,
|
254 |
+
"step": 400
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.69,
|
258 |
+
"eval_distillation_accuracy_counterfactual": 0.5179579271421242,
|
259 |
+
"eval_distillation_accuracy_factual": 0.7955361723961005,
|
260 |
+
"eval_distillation_f1_counterfactual": 0.4694720290698628,
|
261 |
+
"eval_distillation_f1_factual": 0.7787731275919787,
|
262 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5007696254489482,
|
263 |
+
"eval_groundtruth_f1_counterfactual": 0.44241001562042664,
|
264 |
+
"eval_groundtruth_f1_factual": 0.5479747801171813,
|
265 |
+
"eval_icace_cosine": 0.4990290403366089,
|
266 |
+
"eval_icace_l2": 0.6423947811126709,
|
267 |
+
"eval_icace_normdiff": 0.4277788996696472,
|
268 |
+
"eval_loss": 10.05081558227539,
|
269 |
+
"eval_runtime": 4.5234,
|
270 |
+
"eval_samples_per_second": 861.75,
|
271 |
+
"eval_steps_per_second": 6.853,
|
272 |
+
"step": 400
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 0.72,
|
276 |
+
"learning_rate": 4.5450606585788563e-05,
|
277 |
+
"loss": 14.2025,
|
278 |
+
"step": 420
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 0.75,
|
282 |
+
"learning_rate": 4.5233968804159446e-05,
|
283 |
+
"loss": 14.1203,
|
284 |
+
"step": 440
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 0.77,
|
288 |
+
"eval_distillation_accuracy_counterfactual": 0.5023088763468445,
|
289 |
+
"eval_distillation_accuracy_factual": 0.8127244740892765,
|
290 |
+
"eval_distillation_f1_counterfactual": 0.4656964220693382,
|
291 |
+
"eval_distillation_f1_factual": 0.8006616317977919,
|
292 |
+
"eval_groundtruth_accuracy_counterfactual": 0.4789635710620831,
|
293 |
+
"eval_groundtruth_f1_counterfactual": 0.4346768246991469,
|
294 |
+
"eval_groundtruth_f1_factual": 0.5734933636976969,
|
295 |
+
"eval_icace_cosine": 0.5007656216621399,
|
296 |
+
"eval_icace_l2": 0.6549262404441833,
|
297 |
+
"eval_icace_normdiff": 0.4307872951030731,
|
298 |
+
"eval_loss": 10.223078727722168,
|
299 |
+
"eval_runtime": 4.7741,
|
300 |
+
"eval_samples_per_second": 816.492,
|
301 |
+
"eval_steps_per_second": 6.493,
|
302 |
+
"step": 450
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 0.79,
|
306 |
+
"learning_rate": 4.501733102253033e-05,
|
307 |
+
"loss": 14.0055,
|
308 |
+
"step": 460
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 0.82,
|
312 |
+
"learning_rate": 4.480069324090121e-05,
|
313 |
+
"loss": 14.1732,
|
314 |
+
"step": 480
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 0.86,
|
318 |
+
"learning_rate": 4.45840554592721e-05,
|
319 |
+
"loss": 13.7579,
|
320 |
+
"step": 500
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 0.86,
|
324 |
+
"eval_distillation_accuracy_counterfactual": 0.5017957927142124,
|
325 |
+
"eval_distillation_accuracy_factual": 0.8217034376603386,
|
326 |
+
"eval_distillation_f1_counterfactual": 0.45110748338345036,
|
327 |
+
"eval_distillation_f1_factual": 0.8120033725947984,
|
328 |
+
"eval_groundtruth_accuracy_counterfactual": 0.491277578245254,
|
329 |
+
"eval_groundtruth_f1_counterfactual": 0.4306026913746316,
|
330 |
+
"eval_groundtruth_f1_factual": 0.5751269063395339,
|
331 |
+
"eval_icace_cosine": 0.5004109144210815,
|
332 |
+
"eval_icace_l2": 0.6552226543426514,
|
333 |
+
"eval_icace_normdiff": 0.43052297830581665,
|
334 |
+
"eval_loss": 10.176661491394043,
|
335 |
+
"eval_runtime": 8.7551,
|
336 |
+
"eval_samples_per_second": 445.225,
|
337 |
+
"eval_steps_per_second": 3.541,
|
338 |
+
"step": 500
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.89,
|
342 |
+
"learning_rate": 4.436741767764298e-05,
|
343 |
+
"loss": 13.774,
|
344 |
+
"step": 520
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 0.93,
|
348 |
+
"learning_rate": 4.415077989601387e-05,
|
349 |
+
"loss": 13.8,
|
350 |
+
"step": 540
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 0.94,
|
354 |
+
"eval_distillation_accuracy_counterfactual": 0.5059004617752694,
|
355 |
+
"eval_distillation_accuracy_factual": 0.8234992303745511,
|
356 |
+
"eval_distillation_f1_counterfactual": 0.4512025853019249,
|
357 |
+
"eval_distillation_f1_factual": 0.8135629436467913,
|
358 |
+
"eval_groundtruth_accuracy_counterfactual": 0.49461262185736277,
|
359 |
+
"eval_groundtruth_f1_counterfactual": 0.43348852199044147,
|
360 |
+
"eval_groundtruth_f1_factual": 0.5789128560578634,
|
361 |
+
"eval_icace_cosine": 0.498872309923172,
|
362 |
+
"eval_icace_l2": 0.6550716757774353,
|
363 |
+
"eval_icace_normdiff": 0.42937180399894714,
|
364 |
+
"eval_loss": 10.230228424072266,
|
365 |
+
"eval_runtime": 7.836,
|
366 |
+
"eval_samples_per_second": 497.446,
|
367 |
+
"eval_steps_per_second": 3.956,
|
368 |
+
"step": 550
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 0.96,
|
372 |
+
"learning_rate": 4.393414211438475e-05,
|
373 |
+
"loss": 13.5074,
|
374 |
+
"step": 560
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 0.99,
|
378 |
+
"learning_rate": 4.371750433275563e-05,
|
379 |
+
"loss": 13.6928,
|
380 |
+
"step": 580
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 1.03,
|
384 |
+
"learning_rate": 4.3500866551126516e-05,
|
385 |
+
"loss": 13.7141,
|
386 |
+
"step": 600
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 1.03,
|
390 |
+
"eval_distillation_accuracy_counterfactual": 0.5179579271421242,
|
391 |
+
"eval_distillation_accuracy_factual": 0.8306824012314007,
|
392 |
+
"eval_distillation_f1_counterfactual": 0.45833699531570044,
|
393 |
+
"eval_distillation_f1_factual": 0.8222217849303248,
|
394 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5033350436121088,
|
395 |
+
"eval_groundtruth_f1_counterfactual": 0.433570321967066,
|
396 |
+
"eval_groundtruth_f1_factual": 0.5763197755988125,
|
397 |
+
"eval_icace_cosine": 0.49790823459625244,
|
398 |
+
"eval_icace_l2": 0.6532407402992249,
|
399 |
+
"eval_icace_normdiff": 0.4311390519142151,
|
400 |
+
"eval_loss": 10.064295768737793,
|
401 |
+
"eval_runtime": 4.6211,
|
402 |
+
"eval_samples_per_second": 843.531,
|
403 |
+
"eval_steps_per_second": 6.708,
|
404 |
+
"step": 600
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 1.06,
|
408 |
+
"learning_rate": 4.3284228769497406e-05,
|
409 |
+
"loss": 13.3673,
|
410 |
+
"step": 620
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 1.1,
|
414 |
+
"learning_rate": 4.306759098786828e-05,
|
415 |
+
"loss": 13.4799,
|
416 |
+
"step": 640
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 1.11,
|
420 |
+
"eval_distillation_accuracy_counterfactual": 0.5187275525910724,
|
421 |
+
"eval_distillation_accuracy_factual": 0.8283735248845562,
|
422 |
+
"eval_distillation_f1_counterfactual": 0.46382143672624687,
|
423 |
+
"eval_distillation_f1_factual": 0.8207720593293605,
|
424 |
+
"eval_groundtruth_accuracy_counterfactual": 0.5064135454079015,
|
425 |
+
"eval_groundtruth_f1_counterfactual": 0.4409147466060839,
|
426 |
+
"eval_groundtruth_f1_factual": 0.5772188705192105,
|
427 |
+
"eval_icace_cosine": 0.49332907795906067,
|
428 |
+
"eval_icace_l2": 0.6394243240356445,
|
429 |
+
"eval_icace_normdiff": 0.4169619083404541,
|
430 |
+
"eval_loss": 9.886384963989258,
|
431 |
+
"eval_runtime": 4.5468,
|
432 |
+
"eval_samples_per_second": 857.313,
|
433 |
+
"eval_steps_per_second": 6.818,
|
434 |
+
"step": 650
|
435 |
+
}
|
436 |
+
],
|
437 |
+
"max_steps": 4616,
|
438 |
+
"num_train_epochs": 8,
|
439 |
+
"total_flos": 98183903155200.0,
|
440 |
+
"trial_name": null,
|
441 |
+
"trial_params": null
|
442 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8d7113dc9437f3e3f21903b115446f62fa83b29c6107184a0802cd28da95a05a
|
3 |
+
size 3439
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|