first commit
Browse files- config.json +26 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- trainer_state.json +632 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "voidism/diffcse-roberta-base-trans",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaModel"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"gradient_checkpointing": false,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "roberta",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"transformers_version": "4.2.1",
|
| 23 |
+
"type_vocab_size": 1,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 50265
|
| 26 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:887994f8b60a5e72c2b3063b05472612ef439c3862d0f887262296a8775e206c
|
| 3 |
+
size 1487809454
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "special_tokens_map_file": null, "name_or_path": "./roberta-base"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,632 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8359799918026352,
|
| 3 |
+
"best_model_checkpoint": "result/simcse-celectra-amlp-dmlp-bs128-lr2e-6-mask0.40-elew0.01-roberta-base",
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"global_step": 7813,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"electra_acc": 0.2448,
|
| 12 |
+
"electra_fix_acc": 0.006,
|
| 13 |
+
"electra_rep_acc": 0.9912,
|
| 14 |
+
"epoch": 0.0,
|
| 15 |
+
"learning_rate": 1.9997440163829513e-06,
|
| 16 |
+
"loss": 8.6888,
|
| 17 |
+
"neg_sim": 0.2041,
|
| 18 |
+
"pos_sim": 0.3651,
|
| 19 |
+
"step": 1
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.02,
|
| 23 |
+
"eval_avg_sts": 0.7368351609735859,
|
| 24 |
+
"eval_sickr_spearman": 0.6964049570574558,
|
| 25 |
+
"eval_stsb_spearman": 0.7772653648897162,
|
| 26 |
+
"step": 125
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.03,
|
| 30 |
+
"eval_avg_sts": 0.7530561409181189,
|
| 31 |
+
"eval_sickr_spearman": 0.707297882521504,
|
| 32 |
+
"eval_stsb_spearman": 0.7988143993147339,
|
| 33 |
+
"step": 250
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.05,
|
| 37 |
+
"eval_avg_sts": 0.7605110871922394,
|
| 38 |
+
"eval_sickr_spearman": 0.7124740982612865,
|
| 39 |
+
"eval_stsb_spearman": 0.8085480761231921,
|
| 40 |
+
"step": 375
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"electra_acc": 0.7243,
|
| 44 |
+
"electra_fix_acc": 0.8981,
|
| 45 |
+
"electra_rep_acc": 0.2064,
|
| 46 |
+
"epoch": 0.06,
|
| 47 |
+
"learning_rate": 1.8720081914757456e-06,
|
| 48 |
+
"loss": 0.6062,
|
| 49 |
+
"neg_sim": 0.0055,
|
| 50 |
+
"pos_sim": 0.6593,
|
| 51 |
+
"step": 500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.06,
|
| 55 |
+
"eval_avg_sts": 0.763939631028596,
|
| 56 |
+
"eval_sickr_spearman": 0.7155249377624325,
|
| 57 |
+
"eval_stsb_spearman": 0.8123543242947595,
|
| 58 |
+
"step": 500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.08,
|
| 62 |
+
"eval_avg_sts": 0.7670128200581756,
|
| 63 |
+
"eval_sickr_spearman": 0.7176028112417467,
|
| 64 |
+
"eval_stsb_spearman": 0.8164228288746044,
|
| 65 |
+
"step": 625
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1,
|
| 69 |
+
"eval_avg_sts": 0.7683647856842559,
|
| 70 |
+
"eval_sickr_spearman": 0.7182528161375087,
|
| 71 |
+
"eval_stsb_spearman": 0.8184767552310033,
|
| 72 |
+
"step": 750
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.11,
|
| 76 |
+
"eval_avg_sts": 0.7703405632893722,
|
| 77 |
+
"eval_sickr_spearman": 0.7201300637061159,
|
| 78 |
+
"eval_stsb_spearman": 0.8205510628726285,
|
| 79 |
+
"step": 875
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"electra_acc": 0.7954,
|
| 83 |
+
"electra_fix_acc": 0.9411,
|
| 84 |
+
"electra_rep_acc": 0.3621,
|
| 85 |
+
"epoch": 0.13,
|
| 86 |
+
"learning_rate": 1.744016382951491e-06,
|
| 87 |
+
"loss": 0.0043,
|
| 88 |
+
"neg_sim": -0.0066,
|
| 89 |
+
"pos_sim": 0.7176,
|
| 90 |
+
"step": 1000
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.13,
|
| 94 |
+
"eval_avg_sts": 0.7716371834988085,
|
| 95 |
+
"eval_sickr_spearman": 0.7207027865596639,
|
| 96 |
+
"eval_stsb_spearman": 0.8225715804379532,
|
| 97 |
+
"step": 1000
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.14,
|
| 101 |
+
"eval_avg_sts": 0.7719569589534554,
|
| 102 |
+
"eval_sickr_spearman": 0.7195236710504597,
|
| 103 |
+
"eval_stsb_spearman": 0.8243902468564511,
|
| 104 |
+
"step": 1125
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.16,
|
| 108 |
+
"eval_avg_sts": 0.7725905596083784,
|
| 109 |
+
"eval_sickr_spearman": 0.7204897205936844,
|
| 110 |
+
"eval_stsb_spearman": 0.8246913986230725,
|
| 111 |
+
"step": 1250
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.18,
|
| 115 |
+
"eval_avg_sts": 0.7740128875580021,
|
| 116 |
+
"eval_sickr_spearman": 0.7209413090094056,
|
| 117 |
+
"eval_stsb_spearman": 0.8270844661065986,
|
| 118 |
+
"step": 1375
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"electra_acc": 0.806,
|
| 122 |
+
"electra_fix_acc": 0.9446,
|
| 123 |
+
"electra_rep_acc": 0.3921,
|
| 124 |
+
"epoch": 0.19,
|
| 125 |
+
"learning_rate": 1.6160245744272365e-06,
|
| 126 |
+
"loss": 0.0039,
|
| 127 |
+
"neg_sim": -0.0067,
|
| 128 |
+
"pos_sim": 0.7288,
|
| 129 |
+
"step": 1500
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.19,
|
| 133 |
+
"eval_avg_sts": 0.7747104419219604,
|
| 134 |
+
"eval_sickr_spearman": 0.721527768757965,
|
| 135 |
+
"eval_stsb_spearman": 0.8278931150859556,
|
| 136 |
+
"step": 1500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.21,
|
| 140 |
+
"eval_avg_sts": 0.7750669451240817,
|
| 141 |
+
"eval_sickr_spearman": 0.722242135329654,
|
| 142 |
+
"eval_stsb_spearman": 0.8278917549185095,
|
| 143 |
+
"step": 1625
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.22,
|
| 147 |
+
"eval_avg_sts": 0.7751917311744843,
|
| 148 |
+
"eval_sickr_spearman": 0.7214598527805315,
|
| 149 |
+
"eval_stsb_spearman": 0.8289236095684369,
|
| 150 |
+
"step": 1750
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.24,
|
| 154 |
+
"eval_avg_sts": 0.775110379872699,
|
| 155 |
+
"eval_sickr_spearman": 0.7211095139266419,
|
| 156 |
+
"eval_stsb_spearman": 0.8291112458187563,
|
| 157 |
+
"step": 1875
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"electra_acc": 0.8111,
|
| 161 |
+
"electra_fix_acc": 0.9455,
|
| 162 |
+
"electra_rep_acc": 0.4077,
|
| 163 |
+
"epoch": 0.26,
|
| 164 |
+
"learning_rate": 1.4880327659029822e-06,
|
| 165 |
+
"loss": 0.0037,
|
| 166 |
+
"neg_sim": -0.0067,
|
| 167 |
+
"pos_sim": 0.7361,
|
| 168 |
+
"step": 2000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.26,
|
| 172 |
+
"eval_avg_sts": 0.7752389174925427,
|
| 173 |
+
"eval_sickr_spearman": 0.7207011535022151,
|
| 174 |
+
"eval_stsb_spearman": 0.8297766814828704,
|
| 175 |
+
"step": 2000
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.27,
|
| 179 |
+
"eval_avg_sts": 0.7759291537103485,
|
| 180 |
+
"eval_sickr_spearman": 0.7218097593545081,
|
| 181 |
+
"eval_stsb_spearman": 0.8300485480661889,
|
| 182 |
+
"step": 2125
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.29,
|
| 186 |
+
"eval_avg_sts": 0.7759100192724719,
|
| 187 |
+
"eval_sickr_spearman": 0.7214468843831432,
|
| 188 |
+
"eval_stsb_spearman": 0.8303731541618006,
|
| 189 |
+
"step": 2250
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.3,
|
| 193 |
+
"eval_avg_sts": 0.77580052054564,
|
| 194 |
+
"eval_sickr_spearman": 0.7208077825474077,
|
| 195 |
+
"eval_stsb_spearman": 0.8307932585438723,
|
| 196 |
+
"step": 2375
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"electra_acc": 0.8137,
|
| 200 |
+
"electra_fix_acc": 0.9463,
|
| 201 |
+
"electra_rep_acc": 0.4184,
|
| 202 |
+
"epoch": 0.32,
|
| 203 |
+
"learning_rate": 1.3600409573787276e-06,
|
| 204 |
+
"loss": 0.0036,
|
| 205 |
+
"neg_sim": -0.0067,
|
| 206 |
+
"pos_sim": 0.7422,
|
| 207 |
+
"step": 2500
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.32,
|
| 211 |
+
"eval_avg_sts": 0.7764057696201115,
|
| 212 |
+
"eval_sickr_spearman": 0.7207564372999703,
|
| 213 |
+
"eval_stsb_spearman": 0.8320551019402528,
|
| 214 |
+
"step": 2500
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.34,
|
| 218 |
+
"eval_avg_sts": 0.7771167451325478,
|
| 219 |
+
"eval_sickr_spearman": 0.7216867036726237,
|
| 220 |
+
"eval_stsb_spearman": 0.8325467865924719,
|
| 221 |
+
"step": 2625
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.35,
|
| 225 |
+
"eval_avg_sts": 0.777409082133846,
|
| 226 |
+
"eval_sickr_spearman": 0.7218441496231378,
|
| 227 |
+
"eval_stsb_spearman": 0.8329740146445542,
|
| 228 |
+
"step": 2750
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.37,
|
| 232 |
+
"eval_avg_sts": 0.7776325118235485,
|
| 233 |
+
"eval_sickr_spearman": 0.722264277667417,
|
| 234 |
+
"eval_stsb_spearman": 0.8330007459796801,
|
| 235 |
+
"step": 2875
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"electra_acc": 0.8163,
|
| 239 |
+
"electra_fix_acc": 0.9464,
|
| 240 |
+
"electra_rep_acc": 0.4284,
|
| 241 |
+
"epoch": 0.38,
|
| 242 |
+
"learning_rate": 1.2320491488544733e-06,
|
| 243 |
+
"loss": 0.0034,
|
| 244 |
+
"neg_sim": -0.0068,
|
| 245 |
+
"pos_sim": 0.7471,
|
| 246 |
+
"step": 3000
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.38,
|
| 250 |
+
"eval_avg_sts": 0.7777138423021386,
|
| 251 |
+
"eval_sickr_spearman": 0.7227461256770441,
|
| 252 |
+
"eval_stsb_spearman": 0.832681558927233,
|
| 253 |
+
"step": 3000
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.4,
|
| 257 |
+
"eval_avg_sts": 0.7781702830872564,
|
| 258 |
+
"eval_sickr_spearman": 0.7235760070476919,
|
| 259 |
+
"eval_stsb_spearman": 0.832764559126821,
|
| 260 |
+
"step": 3125
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.42,
|
| 264 |
+
"eval_avg_sts": 0.7788742472264717,
|
| 265 |
+
"eval_sickr_spearman": 0.7235008864050426,
|
| 266 |
+
"eval_stsb_spearman": 0.8342476080479008,
|
| 267 |
+
"step": 3250
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.43,
|
| 271 |
+
"eval_avg_sts": 0.7786540883853343,
|
| 272 |
+
"eval_sickr_spearman": 0.7228896906392426,
|
| 273 |
+
"eval_stsb_spearman": 0.8344184861314262,
|
| 274 |
+
"step": 3375
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"electra_acc": 0.8185,
|
| 278 |
+
"electra_fix_acc": 0.9468,
|
| 279 |
+
"electra_rep_acc": 0.4365,
|
| 280 |
+
"epoch": 0.45,
|
| 281 |
+
"learning_rate": 1.104057340330219e-06,
|
| 282 |
+
"loss": 0.0034,
|
| 283 |
+
"neg_sim": -0.0068,
|
| 284 |
+
"pos_sim": 0.7518,
|
| 285 |
+
"step": 3500
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.45,
|
| 289 |
+
"eval_avg_sts": 0.7785390972296262,
|
| 290 |
+
"eval_sickr_spearman": 0.7228380572051966,
|
| 291 |
+
"eval_stsb_spearman": 0.8342401372540559,
|
| 292 |
+
"step": 3500
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.46,
|
| 296 |
+
"eval_avg_sts": 0.778413022922875,
|
| 297 |
+
"eval_sickr_spearman": 0.7219714320419488,
|
| 298 |
+
"eval_stsb_spearman": 0.8348546138038011,
|
| 299 |
+
"step": 3625
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.48,
|
| 303 |
+
"eval_avg_sts": 0.778801601085773,
|
| 304 |
+
"eval_sickr_spearman": 0.722748959512029,
|
| 305 |
+
"eval_stsb_spearman": 0.8348542426595171,
|
| 306 |
+
"step": 3750
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.5,
|
| 310 |
+
"eval_avg_sts": 0.7782529644249387,
|
| 311 |
+
"eval_sickr_spearman": 0.7217000082877221,
|
| 312 |
+
"eval_stsb_spearman": 0.8348059205621555,
|
| 313 |
+
"step": 3875
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"electra_acc": 0.8202,
|
| 317 |
+
"electra_fix_acc": 0.9468,
|
| 318 |
+
"electra_rep_acc": 0.4426,
|
| 319 |
+
"epoch": 0.51,
|
| 320 |
+
"learning_rate": 9.760655318059644e-07,
|
| 321 |
+
"loss": 0.0033,
|
| 322 |
+
"neg_sim": -0.0068,
|
| 323 |
+
"pos_sim": 0.7559,
|
| 324 |
+
"step": 4000
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.51,
|
| 328 |
+
"eval_avg_sts": 0.7782394657347366,
|
| 329 |
+
"eval_sickr_spearman": 0.7217301237883238,
|
| 330 |
+
"eval_stsb_spearman": 0.8347488076811493,
|
| 331 |
+
"step": 4000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.53,
|
| 335 |
+
"eval_avg_sts": 0.7782895079814026,
|
| 336 |
+
"eval_sickr_spearman": 0.7220120663537655,
|
| 337 |
+
"eval_stsb_spearman": 0.8345669496090398,
|
| 338 |
+
"step": 4125
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.54,
|
| 342 |
+
"eval_avg_sts": 0.778080439519264,
|
| 343 |
+
"eval_sickr_spearman": 0.7216004398144409,
|
| 344 |
+
"eval_stsb_spearman": 0.834560439224087,
|
| 345 |
+
"step": 4250
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.56,
|
| 349 |
+
"eval_avg_sts": 0.7782284276808015,
|
| 350 |
+
"eval_sickr_spearman": 0.7216616794687745,
|
| 351 |
+
"eval_stsb_spearman": 0.8347951758928285,
|
| 352 |
+
"step": 4375
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"electra_acc": 0.8217,
|
| 356 |
+
"electra_fix_acc": 0.9477,
|
| 357 |
+
"electra_rep_acc": 0.4458,
|
| 358 |
+
"epoch": 0.58,
|
| 359 |
+
"learning_rate": 8.480737232817099e-07,
|
| 360 |
+
"loss": 0.0032,
|
| 361 |
+
"neg_sim": -0.0068,
|
| 362 |
+
"pos_sim": 0.7596,
|
| 363 |
+
"step": 4500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.58,
|
| 367 |
+
"eval_avg_sts": 0.7781400153641456,
|
| 368 |
+
"eval_sickr_spearman": 0.7214844927355691,
|
| 369 |
+
"eval_stsb_spearman": 0.8347955379927221,
|
| 370 |
+
"step": 4500
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.59,
|
| 374 |
+
"eval_avg_sts": 0.7783651714396398,
|
| 375 |
+
"eval_sickr_spearman": 0.721677865949959,
|
| 376 |
+
"eval_stsb_spearman": 0.8350524769293207,
|
| 377 |
+
"step": 4625
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.61,
|
| 381 |
+
"eval_avg_sts": 0.7779810461862591,
|
| 382 |
+
"eval_sickr_spearman": 0.7215376151337598,
|
| 383 |
+
"eval_stsb_spearman": 0.8344244772387583,
|
| 384 |
+
"step": 4750
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.62,
|
| 388 |
+
"eval_avg_sts": 0.7774863632137369,
|
| 389 |
+
"eval_sickr_spearman": 0.7212506225972007,
|
| 390 |
+
"eval_stsb_spearman": 0.833722103830273,
|
| 391 |
+
"step": 4875
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"electra_acc": 0.8222,
|
| 395 |
+
"electra_fix_acc": 0.9476,
|
| 396 |
+
"electra_rep_acc": 0.4478,
|
| 397 |
+
"epoch": 0.64,
|
| 398 |
+
"learning_rate": 7.200819147574555e-07,
|
| 399 |
+
"loss": 0.0033,
|
| 400 |
+
"neg_sim": -0.0068,
|
| 401 |
+
"pos_sim": 0.7614,
|
| 402 |
+
"step": 5000
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.64,
|
| 406 |
+
"eval_avg_sts": 0.7780441928626822,
|
| 407 |
+
"eval_sickr_spearman": 0.7222703776172996,
|
| 408 |
+
"eval_stsb_spearman": 0.833818008108065,
|
| 409 |
+
"step": 5000
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.66,
|
| 413 |
+
"eval_avg_sts": 0.7772624437857663,
|
| 414 |
+
"eval_sickr_spearman": 0.7211198886445525,
|
| 415 |
+
"eval_stsb_spearman": 0.8334049989269801,
|
| 416 |
+
"step": 5125
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.67,
|
| 420 |
+
"eval_avg_sts": 0.777433289264948,
|
| 421 |
+
"eval_sickr_spearman": 0.7208792047952461,
|
| 422 |
+
"eval_stsb_spearman": 0.83398737373465,
|
| 423 |
+
"step": 5250
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.69,
|
| 427 |
+
"eval_avg_sts": 0.778471716959233,
|
| 428 |
+
"eval_sickr_spearman": 0.7219667730251094,
|
| 429 |
+
"eval_stsb_spearman": 0.8349766608933566,
|
| 430 |
+
"step": 5375
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"electra_acc": 0.824,
|
| 434 |
+
"electra_fix_acc": 0.9479,
|
| 435 |
+
"electra_rep_acc": 0.4538,
|
| 436 |
+
"epoch": 0.7,
|
| 437 |
+
"learning_rate": 5.920901062332011e-07,
|
| 438 |
+
"loss": 0.0033,
|
| 439 |
+
"neg_sim": -0.0068,
|
| 440 |
+
"pos_sim": 0.7634,
|
| 441 |
+
"step": 5500
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.7,
|
| 445 |
+
"eval_avg_sts": 0.7783835165868276,
|
| 446 |
+
"eval_sickr_spearman": 0.7220774366828226,
|
| 447 |
+
"eval_stsb_spearman": 0.8346895964908326,
|
| 448 |
+
"step": 5500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.72,
|
| 452 |
+
"eval_avg_sts": 0.7777215174417764,
|
| 453 |
+
"eval_sickr_spearman": 0.7211491395853283,
|
| 454 |
+
"eval_stsb_spearman": 0.8342938952982245,
|
| 455 |
+
"step": 5625
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 0.74,
|
| 459 |
+
"eval_avg_sts": 0.7783006770868415,
|
| 460 |
+
"eval_sickr_spearman": 0.7218946783418506,
|
| 461 |
+
"eval_stsb_spearman": 0.8347066758318323,
|
| 462 |
+
"step": 5750
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 0.75,
|
| 466 |
+
"eval_avg_sts": 0.7786525912567952,
|
| 467 |
+
"eval_sickr_spearman": 0.7218885303608666,
|
| 468 |
+
"eval_stsb_spearman": 0.8354166521527239,
|
| 469 |
+
"step": 5875
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"electra_acc": 0.8245,
|
| 473 |
+
"electra_fix_acc": 0.9479,
|
| 474 |
+
"electra_rep_acc": 0.4557,
|
| 475 |
+
"epoch": 0.77,
|
| 476 |
+
"learning_rate": 4.640982977089466e-07,
|
| 477 |
+
"loss": 0.0033,
|
| 478 |
+
"neg_sim": -0.0068,
|
| 479 |
+
"pos_sim": 0.7639,
|
| 480 |
+
"step": 6000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.77,
|
| 484 |
+
"eval_avg_sts": 0.7786598191346372,
|
| 485 |
+
"eval_sickr_spearman": 0.7214591323140099,
|
| 486 |
+
"eval_stsb_spearman": 0.8358605059552645,
|
| 487 |
+
"step": 6000
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.78,
|
| 491 |
+
"eval_avg_sts": 0.7785391387872032,
|
| 492 |
+
"eval_sickr_spearman": 0.7212359798167284,
|
| 493 |
+
"eval_stsb_spearman": 0.8358422977576779,
|
| 494 |
+
"step": 6125
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.8,
|
| 498 |
+
"eval_avg_sts": 0.7780962443354171,
|
| 499 |
+
"eval_sickr_spearman": 0.7207366965172793,
|
| 500 |
+
"eval_stsb_spearman": 0.835455792153555,
|
| 501 |
+
"step": 6250
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.82,
|
| 505 |
+
"eval_avg_sts": 0.7787170644410477,
|
| 506 |
+
"eval_sickr_spearman": 0.7214541370794603,
|
| 507 |
+
"eval_stsb_spearman": 0.8359799918026352,
|
| 508 |
+
"step": 6375
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"electra_acc": 0.8249,
|
| 512 |
+
"electra_fix_acc": 0.9477,
|
| 513 |
+
"electra_rep_acc": 0.458,
|
| 514 |
+
"epoch": 0.83,
|
| 515 |
+
"learning_rate": 3.361064891846921e-07,
|
| 516 |
+
"loss": 0.0034,
|
| 517 |
+
"neg_sim": -0.0068,
|
| 518 |
+
"pos_sim": 0.765,
|
| 519 |
+
"step": 6500
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.83,
|
| 523 |
+
"eval_avg_sts": 0.7787359930993909,
|
| 524 |
+
"eval_sickr_spearman": 0.7215321395881958,
|
| 525 |
+
"eval_stsb_spearman": 0.835939846610586,
|
| 526 |
+
"step": 6500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.85,
|
| 530 |
+
"eval_avg_sts": 0.778662230635321,
|
| 531 |
+
"eval_sickr_spearman": 0.7217551960232744,
|
| 532 |
+
"eval_stsb_spearman": 0.8355692652473675,
|
| 533 |
+
"step": 6625
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 0.86,
|
| 537 |
+
"eval_avg_sts": 0.7788844288300487,
|
| 538 |
+
"eval_sickr_spearman": 0.7221871330355174,
|
| 539 |
+
"eval_stsb_spearman": 0.8355817246245801,
|
| 540 |
+
"step": 6750
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 0.88,
|
| 544 |
+
"eval_avg_sts": 0.7789522767655956,
|
| 545 |
+
"eval_sickr_spearman": 0.722305728507958,
|
| 546 |
+
"eval_stsb_spearman": 0.8355988250232333,
|
| 547 |
+
"step": 6875
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"electra_acc": 0.8251,
|
| 551 |
+
"electra_fix_acc": 0.9476,
|
| 552 |
+
"electra_rep_acc": 0.4589,
|
| 553 |
+
"epoch": 0.9,
|
| 554 |
+
"learning_rate": 2.0811468066043772e-07,
|
| 555 |
+
"loss": 0.0034,
|
| 556 |
+
"neg_sim": -0.0069,
|
| 557 |
+
"pos_sim": 0.7655,
|
| 558 |
+
"step": 7000
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 0.9,
|
| 562 |
+
"eval_avg_sts": 0.7787617595098588,
|
| 563 |
+
"eval_sickr_spearman": 0.721966628931805,
|
| 564 |
+
"eval_stsb_spearman": 0.8355568900879128,
|
| 565 |
+
"step": 7000
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 0.91,
|
| 569 |
+
"eval_avg_sts": 0.7786478565950568,
|
| 570 |
+
"eval_sickr_spearman": 0.7216480386359659,
|
| 571 |
+
"eval_stsb_spearman": 0.8356476745541477,
|
| 572 |
+
"step": 7125
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.93,
|
| 576 |
+
"eval_avg_sts": 0.7787981847358625,
|
| 577 |
+
"eval_sickr_spearman": 0.7219965523080009,
|
| 578 |
+
"eval_stsb_spearman": 0.8355998171637242,
|
| 579 |
+
"step": 7250
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.94,
|
| 583 |
+
"eval_avg_sts": 0.7788565954562792,
|
| 584 |
+
"eval_sickr_spearman": 0.7221892050558691,
|
| 585 |
+
"eval_stsb_spearman": 0.8355239858566892,
|
| 586 |
+
"step": 7375
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"electra_acc": 0.8254,
|
| 590 |
+
"electra_fix_acc": 0.9477,
|
| 591 |
+
"electra_rep_acc": 0.4607,
|
| 592 |
+
"epoch": 0.96,
|
| 593 |
+
"learning_rate": 8.012287213618328e-08,
|
| 594 |
+
"loss": 0.0032,
|
| 595 |
+
"neg_sim": -0.0069,
|
| 596 |
+
"pos_sim": 0.7665,
|
| 597 |
+
"step": 7500
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.96,
|
| 601 |
+
"eval_avg_sts": 0.7788488501171182,
|
| 602 |
+
"eval_sickr_spearman": 0.7221623076390639,
|
| 603 |
+
"eval_stsb_spearman": 0.8355353925951726,
|
| 604 |
+
"step": 7500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.98,
|
| 608 |
+
"eval_avg_sts": 0.7788989946596425,
|
| 609 |
+
"eval_sickr_spearman": 0.7222944411991201,
|
| 610 |
+
"eval_stsb_spearman": 0.8355035481201649,
|
| 611 |
+
"step": 7625
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.99,
|
| 615 |
+
"eval_avg_sts": 0.7788451325075824,
|
| 616 |
+
"eval_sickr_spearman": 0.7221886767137534,
|
| 617 |
+
"eval_stsb_spearman": 0.8355015883014113,
|
| 618 |
+
"step": 7750
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.0,
|
| 622 |
+
"step": 7813,
|
| 623 |
+
"train_runtime": 7033.0789,
|
| 624 |
+
"train_samples_per_second": 1.111
|
| 625 |
+
}
|
| 626 |
+
],
|
| 627 |
+
"max_steps": 7813,
|
| 628 |
+
"num_train_epochs": 1,
|
| 629 |
+
"total_flos": 285611882411596800,
|
| 630 |
+
"trial_name": null,
|
| 631 |
+
"trial_params": null
|
| 632 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|