root
commited on
Commit
·
a2a2999
1
Parent(s):
aa258a4
init
Browse files- config.json +26 -0
- eval_results.txt +12 -0
- merges.txt +0 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- train_results.txt +3 -0
- trainer_state.json +546 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "result/my-unsup-simcse-roberta-large-d0.1-rs192-std0.5-t0.05",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForCL"
|
| 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": 1024,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 4096,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "roberta",
|
| 18 |
+
"num_attention_heads": 16,
|
| 19 |
+
"num_hidden_layers": 24,
|
| 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 |
+
}
|
eval_results.txt
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
epoch = 1.0
|
| 2 |
+
eval_CR = 87.64
|
| 3 |
+
eval_MPQA = 86.43
|
| 4 |
+
eval_MR = 81.37
|
| 5 |
+
eval_MRPC = 75.47
|
| 6 |
+
eval_SST2 = 86.35
|
| 7 |
+
eval_SUBJ = 92.98
|
| 8 |
+
eval_TREC = 77.75
|
| 9 |
+
eval_avg_sts = 0.8057468354082986
|
| 10 |
+
eval_avg_transfer = 83.99857142857142
|
| 11 |
+
eval_sickr_spearman = 0.7553807134558654
|
| 12 |
+
eval_stsb_spearman = 0.8561129573607319
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e51ae3c6336864f1604dc2d8d77d2198cbfbf10b117af85ca57088688c522bfc
|
| 3 |
+
size 2843123565
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca44ca6c527de40219780aadb622b9ddbc8b37a6b016ae79972fc9424a7d7b71
|
| 3 |
+
size 1425811017
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f5361c9880f42ef10c4484bd35497368d1e9ecc00559160f6a7d27dd47547be4
|
| 3 |
+
size 623
|
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>", "model_max_length": 512, "name_or_path": "roberta-large"}
|
train_results.txt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
epoch = 1.0
|
| 2 |
+
train_runtime = 6281.3522
|
| 3 |
+
train_samples_per_second = 1.244
|
trainer_state.json
ADDED
|
@@ -0,0 +1,546 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8561129573607319,
|
| 3 |
+
"best_model_checkpoint": "result/my-unsup-simcse-roberta-large-d0.1-rs192-std0.5-t0.05",
|
| 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 |
+
"epoch": 0.02,
|
| 12 |
+
"eval_avg_sts": 0.778341475975213,
|
| 13 |
+
"eval_sickr_spearman": 0.7418720622385991,
|
| 14 |
+
"eval_stsb_spearman": 0.8148108897118267,
|
| 15 |
+
"step": 125
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.03,
|
| 19 |
+
"eval_avg_sts": 0.7780892042897212,
|
| 20 |
+
"eval_sickr_spearman": 0.7348078399634872,
|
| 21 |
+
"eval_stsb_spearman": 0.8213705686159553,
|
| 22 |
+
"step": 250
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 0.05,
|
| 26 |
+
"eval_avg_sts": 0.7907877208360297,
|
| 27 |
+
"eval_sickr_spearman": 0.7455401974554278,
|
| 28 |
+
"eval_stsb_spearman": 0.8360352442166316,
|
| 29 |
+
"step": 375
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.06,
|
| 33 |
+
"learning_rate": 1.872008191475746e-05,
|
| 34 |
+
"loss": 0.1943,
|
| 35 |
+
"step": 500
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.06,
|
| 39 |
+
"eval_avg_sts": 0.7955144537705907,
|
| 40 |
+
"eval_sickr_spearman": 0.753126085523259,
|
| 41 |
+
"eval_stsb_spearman": 0.8379028220179224,
|
| 42 |
+
"step": 500
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.08,
|
| 46 |
+
"eval_avg_sts": 0.7917299175053137,
|
| 47 |
+
"eval_sickr_spearman": 0.747866055511467,
|
| 48 |
+
"eval_stsb_spearman": 0.8355937794991605,
|
| 49 |
+
"step": 625
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.1,
|
| 53 |
+
"eval_avg_sts": 0.8016064223978117,
|
| 54 |
+
"eval_sickr_spearman": 0.7573344745690636,
|
| 55 |
+
"eval_stsb_spearman": 0.8458783702265597,
|
| 56 |
+
"step": 750
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.11,
|
| 60 |
+
"eval_avg_sts": 0.8037483952735749,
|
| 61 |
+
"eval_sickr_spearman": 0.7601063676756831,
|
| 62 |
+
"eval_stsb_spearman": 0.8473904228714668,
|
| 63 |
+
"step": 875
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.13,
|
| 67 |
+
"learning_rate": 1.7440163829514915e-05,
|
| 68 |
+
"loss": 0.0005,
|
| 69 |
+
"step": 1000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.13,
|
| 73 |
+
"eval_avg_sts": 0.7957627833498355,
|
| 74 |
+
"eval_sickr_spearman": 0.7503112228234783,
|
| 75 |
+
"eval_stsb_spearman": 0.8412143438761926,
|
| 76 |
+
"step": 1000
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.14,
|
| 80 |
+
"eval_avg_sts": 0.7961233391304097,
|
| 81 |
+
"eval_sickr_spearman": 0.7518578723208871,
|
| 82 |
+
"eval_stsb_spearman": 0.8403888059399321,
|
| 83 |
+
"step": 1125
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.16,
|
| 87 |
+
"eval_avg_sts": 0.8051272481365592,
|
| 88 |
+
"eval_sickr_spearman": 0.7633100719300806,
|
| 89 |
+
"eval_stsb_spearman": 0.8469444243430378,
|
| 90 |
+
"step": 1250
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.18,
|
| 94 |
+
"eval_avg_sts": 0.8057468354082986,
|
| 95 |
+
"eval_sickr_spearman": 0.7553807134558654,
|
| 96 |
+
"eval_stsb_spearman": 0.8561129573607319,
|
| 97 |
+
"step": 1375
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.19,
|
| 101 |
+
"learning_rate": 1.616024574427237e-05,
|
| 102 |
+
"loss": 0.0004,
|
| 103 |
+
"step": 1500
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.19,
|
| 107 |
+
"eval_avg_sts": 0.8123723555741679,
|
| 108 |
+
"eval_sickr_spearman": 0.7723355962013163,
|
| 109 |
+
"eval_stsb_spearman": 0.8524091149470194,
|
| 110 |
+
"step": 1500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.21,
|
| 114 |
+
"eval_avg_sts": 0.8035895611478845,
|
| 115 |
+
"eval_sickr_spearman": 0.7606879100092697,
|
| 116 |
+
"eval_stsb_spearman": 0.8464912122864992,
|
| 117 |
+
"step": 1625
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.22,
|
| 121 |
+
"eval_avg_sts": 0.7991717974519061,
|
| 122 |
+
"eval_sickr_spearman": 0.7465859786270398,
|
| 123 |
+
"eval_stsb_spearman": 0.8517576162767726,
|
| 124 |
+
"step": 1750
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 0.24,
|
| 128 |
+
"eval_avg_sts": 0.7964600696271789,
|
| 129 |
+
"eval_sickr_spearman": 0.7422818155649676,
|
| 130 |
+
"eval_stsb_spearman": 0.8506383236893902,
|
| 131 |
+
"step": 1875
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.26,
|
| 135 |
+
"learning_rate": 1.4880327659029823e-05,
|
| 136 |
+
"loss": 0.0007,
|
| 137 |
+
"step": 2000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.26,
|
| 141 |
+
"eval_avg_sts": 0.8044429784441518,
|
| 142 |
+
"eval_sickr_spearman": 0.7591500982345256,
|
| 143 |
+
"eval_stsb_spearman": 0.8497358586537781,
|
| 144 |
+
"step": 2000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.27,
|
| 148 |
+
"eval_avg_sts": 0.80058268060333,
|
| 149 |
+
"eval_sickr_spearman": 0.7487337853900479,
|
| 150 |
+
"eval_stsb_spearman": 0.8524315758166123,
|
| 151 |
+
"step": 2125
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.29,
|
| 155 |
+
"eval_avg_sts": 0.8019106027671927,
|
| 156 |
+
"eval_sickr_spearman": 0.7506407822258964,
|
| 157 |
+
"eval_stsb_spearman": 0.853180423308489,
|
| 158 |
+
"step": 2250
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.3,
|
| 162 |
+
"eval_avg_sts": 0.8040000758463681,
|
| 163 |
+
"eval_sickr_spearman": 0.7559858573028843,
|
| 164 |
+
"eval_stsb_spearman": 0.852014294389852,
|
| 165 |
+
"step": 2375
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.32,
|
| 169 |
+
"learning_rate": 1.3600409573787278e-05,
|
| 170 |
+
"loss": 0.0004,
|
| 171 |
+
"step": 2500
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.32,
|
| 175 |
+
"eval_avg_sts": 0.8026474310995677,
|
| 176 |
+
"eval_sickr_spearman": 0.7573420154519893,
|
| 177 |
+
"eval_stsb_spearman": 0.847952846747146,
|
| 178 |
+
"step": 2500
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.34,
|
| 182 |
+
"eval_avg_sts": 0.7924131853036531,
|
| 183 |
+
"eval_sickr_spearman": 0.7428461810068655,
|
| 184 |
+
"eval_stsb_spearman": 0.8419801896004406,
|
| 185 |
+
"step": 2625
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.35,
|
| 189 |
+
"eval_avg_sts": 0.7950751062581133,
|
| 190 |
+
"eval_sickr_spearman": 0.7408404502419113,
|
| 191 |
+
"eval_stsb_spearman": 0.8493097622743153,
|
| 192 |
+
"step": 2750
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.37,
|
| 196 |
+
"eval_avg_sts": 0.7999799876999598,
|
| 197 |
+
"eval_sickr_spearman": 0.749456173155677,
|
| 198 |
+
"eval_stsb_spearman": 0.8505038022442425,
|
| 199 |
+
"step": 2875
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.38,
|
| 203 |
+
"learning_rate": 1.2320491488544735e-05,
|
| 204 |
+
"loss": 0.0004,
|
| 205 |
+
"step": 3000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.38,
|
| 209 |
+
"eval_avg_sts": 0.7937458509139534,
|
| 210 |
+
"eval_sickr_spearman": 0.7430877294159979,
|
| 211 |
+
"eval_stsb_spearman": 0.8444039724119089,
|
| 212 |
+
"step": 3000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.4,
|
| 216 |
+
"eval_avg_sts": 0.7856063441760408,
|
| 217 |
+
"eval_sickr_spearman": 0.7330642149190809,
|
| 218 |
+
"eval_stsb_spearman": 0.8381484734330007,
|
| 219 |
+
"step": 3125
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.42,
|
| 223 |
+
"eval_avg_sts": 0.7909319347148922,
|
| 224 |
+
"eval_sickr_spearman": 0.7362177449151016,
|
| 225 |
+
"eval_stsb_spearman": 0.8456461245146829,
|
| 226 |
+
"step": 3250
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.43,
|
| 230 |
+
"eval_avg_sts": 0.7875482454299103,
|
| 231 |
+
"eval_sickr_spearman": 0.7297334021276521,
|
| 232 |
+
"eval_stsb_spearman": 0.8453630887321684,
|
| 233 |
+
"step": 3375
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.45,
|
| 237 |
+
"learning_rate": 1.104057340330219e-05,
|
| 238 |
+
"loss": 0.0004,
|
| 239 |
+
"step": 3500
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.45,
|
| 243 |
+
"eval_avg_sts": 0.7857444621535814,
|
| 244 |
+
"eval_sickr_spearman": 0.7371294712825999,
|
| 245 |
+
"eval_stsb_spearman": 0.8343594530245628,
|
| 246 |
+
"step": 3500
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.46,
|
| 250 |
+
"eval_avg_sts": 0.790497235297885,
|
| 251 |
+
"eval_sickr_spearman": 0.7411372344176974,
|
| 252 |
+
"eval_stsb_spearman": 0.8398572361780724,
|
| 253 |
+
"step": 3625
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.48,
|
| 257 |
+
"eval_avg_sts": 0.7912227836878545,
|
| 258 |
+
"eval_sickr_spearman": 0.7381786146313126,
|
| 259 |
+
"eval_stsb_spearman": 0.8442669527443964,
|
| 260 |
+
"step": 3750
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.5,
|
| 264 |
+
"eval_avg_sts": 0.7960570924578971,
|
| 265 |
+
"eval_sickr_spearman": 0.7468062972893365,
|
| 266 |
+
"eval_stsb_spearman": 0.8453078876264578,
|
| 267 |
+
"step": 3875
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.51,
|
| 271 |
+
"learning_rate": 9.760655318059645e-06,
|
| 272 |
+
"loss": 0.0004,
|
| 273 |
+
"step": 4000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.51,
|
| 277 |
+
"eval_avg_sts": 0.7975689873245354,
|
| 278 |
+
"eval_sickr_spearman": 0.7517150278252102,
|
| 279 |
+
"eval_stsb_spearman": 0.8434229468238607,
|
| 280 |
+
"step": 4000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.53,
|
| 284 |
+
"eval_avg_sts": 0.7971877757425163,
|
| 285 |
+
"eval_sickr_spearman": 0.7490601086932185,
|
| 286 |
+
"eval_stsb_spearman": 0.8453154427918141,
|
| 287 |
+
"step": 4125
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.54,
|
| 291 |
+
"eval_avg_sts": 0.799856091040363,
|
| 292 |
+
"eval_sickr_spearman": 0.7506221281430873,
|
| 293 |
+
"eval_stsb_spearman": 0.8490900539376388,
|
| 294 |
+
"step": 4250
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.56,
|
| 298 |
+
"eval_avg_sts": 0.8009276979255495,
|
| 299 |
+
"eval_sickr_spearman": 0.7507006589939387,
|
| 300 |
+
"eval_stsb_spearman": 0.8511547368571605,
|
| 301 |
+
"step": 4375
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.58,
|
| 305 |
+
"learning_rate": 8.4807372328171e-06,
|
| 306 |
+
"loss": 0.0002,
|
| 307 |
+
"step": 4500
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 0.58,
|
| 311 |
+
"eval_avg_sts": 0.8015427695250972,
|
| 312 |
+
"eval_sickr_spearman": 0.752485398661176,
|
| 313 |
+
"eval_stsb_spearman": 0.8506001403890182,
|
| 314 |
+
"step": 4500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.59,
|
| 318 |
+
"eval_avg_sts": 0.7931466450868088,
|
| 319 |
+
"eval_sickr_spearman": 0.7451727595294263,
|
| 320 |
+
"eval_stsb_spearman": 0.8411205306441913,
|
| 321 |
+
"step": 4625
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.61,
|
| 325 |
+
"eval_avg_sts": 0.8009401798756712,
|
| 326 |
+
"eval_sickr_spearman": 0.7592960647517961,
|
| 327 |
+
"eval_stsb_spearman": 0.8425842949995465,
|
| 328 |
+
"step": 4750
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.62,
|
| 332 |
+
"eval_avg_sts": 0.7902486823800228,
|
| 333 |
+
"eval_sickr_spearman": 0.7485647639440872,
|
| 334 |
+
"eval_stsb_spearman": 0.8319326008159583,
|
| 335 |
+
"step": 4875
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 0.64,
|
| 339 |
+
"learning_rate": 7.200819147574556e-06,
|
| 340 |
+
"loss": 0.0004,
|
| 341 |
+
"step": 5000
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 0.64,
|
| 345 |
+
"eval_avg_sts": 0.7865346213104021,
|
| 346 |
+
"eval_sickr_spearman": 0.7400579275372816,
|
| 347 |
+
"eval_stsb_spearman": 0.8330113150835224,
|
| 348 |
+
"step": 5000
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 0.66,
|
| 352 |
+
"eval_avg_sts": 0.7872787939204771,
|
| 353 |
+
"eval_sickr_spearman": 0.7410168684774935,
|
| 354 |
+
"eval_stsb_spearman": 0.8335407193634606,
|
| 355 |
+
"step": 5125
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 0.67,
|
| 359 |
+
"eval_avg_sts": 0.7881510784701646,
|
| 360 |
+
"eval_sickr_spearman": 0.7505857685992987,
|
| 361 |
+
"eval_stsb_spearman": 0.8257163883410304,
|
| 362 |
+
"step": 5250
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.69,
|
| 366 |
+
"eval_avg_sts": 0.7945145118810497,
|
| 367 |
+
"eval_sickr_spearman": 0.7558576142620445,
|
| 368 |
+
"eval_stsb_spearman": 0.833171409500055,
|
| 369 |
+
"step": 5375
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.7,
|
| 373 |
+
"learning_rate": 5.920901062332012e-06,
|
| 374 |
+
"loss": 0.0002,
|
| 375 |
+
"step": 5500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.7,
|
| 379 |
+
"eval_avg_sts": 0.7979814354987071,
|
| 380 |
+
"eval_sickr_spearman": 0.7559497379146027,
|
| 381 |
+
"eval_stsb_spearman": 0.8400131330828116,
|
| 382 |
+
"step": 5500
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.72,
|
| 386 |
+
"eval_avg_sts": 0.7982480461082944,
|
| 387 |
+
"eval_sickr_spearman": 0.7554273996864633,
|
| 388 |
+
"eval_stsb_spearman": 0.8410686925301255,
|
| 389 |
+
"step": 5625
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.74,
|
| 393 |
+
"eval_avg_sts": 0.798189926599052,
|
| 394 |
+
"eval_sickr_spearman": 0.7587051861419044,
|
| 395 |
+
"eval_stsb_spearman": 0.8376746670561996,
|
| 396 |
+
"step": 5750
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.75,
|
| 400 |
+
"eval_avg_sts": 0.7966011989359767,
|
| 401 |
+
"eval_sickr_spearman": 0.756065973180083,
|
| 402 |
+
"eval_stsb_spearman": 0.8371364246918704,
|
| 403 |
+
"step": 5875
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.77,
|
| 407 |
+
"learning_rate": 4.640982977089467e-06,
|
| 408 |
+
"loss": 0.0004,
|
| 409 |
+
"step": 6000
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.77,
|
| 413 |
+
"eval_avg_sts": 0.7984593901137271,
|
| 414 |
+
"eval_sickr_spearman": 0.7602338720073751,
|
| 415 |
+
"eval_stsb_spearman": 0.8366849082200791,
|
| 416 |
+
"step": 6000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.78,
|
| 420 |
+
"eval_avg_sts": 0.7952611257249883,
|
| 421 |
+
"eval_sickr_spearman": 0.7563395583338746,
|
| 422 |
+
"eval_stsb_spearman": 0.834182693116102,
|
| 423 |
+
"step": 6125
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.8,
|
| 427 |
+
"eval_avg_sts": 0.7946186448265642,
|
| 428 |
+
"eval_sickr_spearman": 0.7567510888109963,
|
| 429 |
+
"eval_stsb_spearman": 0.832486200842132,
|
| 430 |
+
"step": 6250
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.82,
|
| 434 |
+
"eval_avg_sts": 0.7938673811911408,
|
| 435 |
+
"eval_sickr_spearman": 0.756815306393619,
|
| 436 |
+
"eval_stsb_spearman": 0.8309194559886626,
|
| 437 |
+
"step": 6375
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.83,
|
| 441 |
+
"learning_rate": 3.3610648918469217e-06,
|
| 442 |
+
"loss": 0.0003,
|
| 443 |
+
"step": 6500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.83,
|
| 447 |
+
"eval_avg_sts": 0.7941658649824449,
|
| 448 |
+
"eval_sickr_spearman": 0.7582097933616717,
|
| 449 |
+
"eval_stsb_spearman": 0.8301219366032181,
|
| 450 |
+
"step": 6500
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.85,
|
| 454 |
+
"eval_avg_sts": 0.7899228636738962,
|
| 455 |
+
"eval_sickr_spearman": 0.7538527000258148,
|
| 456 |
+
"eval_stsb_spearman": 0.8259930273219777,
|
| 457 |
+
"step": 6625
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.86,
|
| 461 |
+
"eval_avg_sts": 0.7908249597888872,
|
| 462 |
+
"eval_sickr_spearman": 0.7529101376911932,
|
| 463 |
+
"eval_stsb_spearman": 0.8287397818865813,
|
| 464 |
+
"step": 6750
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.88,
|
| 468 |
+
"eval_avg_sts": 0.7913431181790619,
|
| 469 |
+
"eval_sickr_spearman": 0.7534930911693477,
|
| 470 |
+
"eval_stsb_spearman": 0.8291931451887761,
|
| 471 |
+
"step": 6875
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.9,
|
| 475 |
+
"learning_rate": 2.0811468066043775e-06,
|
| 476 |
+
"loss": 0.0004,
|
| 477 |
+
"step": 7000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.9,
|
| 481 |
+
"eval_avg_sts": 0.7932583890535982,
|
| 482 |
+
"eval_sickr_spearman": 0.7548152913297359,
|
| 483 |
+
"eval_stsb_spearman": 0.8317014867774606,
|
| 484 |
+
"step": 7000
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.91,
|
| 488 |
+
"eval_avg_sts": 0.7936961629213847,
|
| 489 |
+
"eval_sickr_spearman": 0.7550173581734861,
|
| 490 |
+
"eval_stsb_spearman": 0.8323749676692832,
|
| 491 |
+
"step": 7125
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.93,
|
| 495 |
+
"eval_avg_sts": 0.7943097969110002,
|
| 496 |
+
"eval_sickr_spearman": 0.7551339296566764,
|
| 497 |
+
"eval_stsb_spearman": 0.8334856641653241,
|
| 498 |
+
"step": 7250
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.94,
|
| 502 |
+
"eval_avg_sts": 0.7952534760489661,
|
| 503 |
+
"eval_sickr_spearman": 0.7558418120296713,
|
| 504 |
+
"eval_stsb_spearman": 0.834665140068261,
|
| 505 |
+
"step": 7375
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.96,
|
| 509 |
+
"learning_rate": 8.012287213618329e-07,
|
| 510 |
+
"loss": 0.0003,
|
| 511 |
+
"step": 7500
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 0.96,
|
| 515 |
+
"eval_avg_sts": 0.7950213976361764,
|
| 516 |
+
"eval_sickr_spearman": 0.7561205845424182,
|
| 517 |
+
"eval_stsb_spearman": 0.8339222107299348,
|
| 518 |
+
"step": 7500
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.98,
|
| 522 |
+
"eval_avg_sts": 0.7959169977411076,
|
| 523 |
+
"eval_sickr_spearman": 0.7571501312017441,
|
| 524 |
+
"eval_stsb_spearman": 0.834683864280471,
|
| 525 |
+
"step": 7625
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 0.99,
|
| 529 |
+
"eval_avg_sts": 0.7962776415425221,
|
| 530 |
+
"eval_sickr_spearman": 0.75766166243206,
|
| 531 |
+
"eval_stsb_spearman": 0.8348936206529841,
|
| 532 |
+
"step": 7750
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 1.0,
|
| 536 |
+
"step": 7813,
|
| 537 |
+
"train_runtime": 6281.3522,
|
| 538 |
+
"train_samples_per_second": 1.244
|
| 539 |
+
}
|
| 540 |
+
],
|
| 541 |
+
"max_steps": 7813,
|
| 542 |
+
"num_train_epochs": 1,
|
| 543 |
+
"total_flos": 136861324957188096,
|
| 544 |
+
"trial_name": null,
|
| 545 |
+
"trial_params": null
|
| 546 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce7c91522135cc4994a6c4306e6ed8c037f953610e2b80b90338200cf2f8f657
|
| 3 |
+
size 2095
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|