Tianyu Gao
commited on
Commit
·
632c976
1
Parent(s):
7331549
init
Browse files- config.json +26 -0
- eval_results.txt +14 -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 +1072 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "result/roberta-base-cls_before_pooler-sym_mlp-mlp_bert-bs64-gpu8-gs1-lr5e-5-m=stsb-cross_contra-norm0.05-l32",
|
| 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": 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 |
+
}
|
eval_results.txt
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
epoch = 3.0
|
| 2 |
+
eval_CR = 92.09
|
| 3 |
+
eval_MPQA = 89.28
|
| 4 |
+
eval_MR = 85.05
|
| 5 |
+
eval_MRPC = 77.31
|
| 6 |
+
eval_SICKEntailment = 84.8
|
| 7 |
+
eval_SST2 = 90.02
|
| 8 |
+
eval_SST5 = 51.32
|
| 9 |
+
eval_SUBJ = 94.44
|
| 10 |
+
eval_TREC = 82.32
|
| 11 |
+
eval_avg_sts = 0.8442180854633452
|
| 12 |
+
eval_avg_transfer = 82.95888888888888
|
| 13 |
+
eval_sickr_spearman = 0.814642927343252
|
| 14 |
+
eval_stsb_spearman = 0.8737932435834382
|
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 sha256:7f061ee5dfd8ae8700a53b83c87e2271d443ec2f3a7cbadbc59eb72eecc82c7b
|
| 3 |
+
size 997286769
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7a542e166481b05379269f684496cd391b891e87f5adb5e4dd80ea6e5df10cb5
|
| 3 |
+
size 498651911
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8b61f8faf9546395d389cb49beeac789bfc6a0ac5766129e037ff2b905c622dd
|
| 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-base"}
|
train_results.txt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
epoch = 3.0
|
| 2 |
+
train_runtime = 1218.5135
|
| 3 |
+
train_samples_per_second = 1.327
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1072 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8737932435834382,
|
| 3 |
+
"best_model_checkpoint": "result/roberta-base-cls_before_pooler-sym_mlp-mlp_bert-bs64-gpu8-gs1-lr5e-5-m=stsb-cross_contra-norm0.05-l32",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"global_step": 1617,
|
| 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 |
+
"learning_rate": 4.969078540507112e-05,
|
| 13 |
+
"loss": 6.1836,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.04,
|
| 18 |
+
"learning_rate": 4.938157081014224e-05,
|
| 19 |
+
"loss": 2.6504,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.06,
|
| 24 |
+
"learning_rate": 4.907235621521336e-05,
|
| 25 |
+
"loss": 1.6665,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.07,
|
| 30 |
+
"learning_rate": 4.876314162028448e-05,
|
| 31 |
+
"loss": 1.3732,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.09,
|
| 36 |
+
"learning_rate": 4.84539270253556e-05,
|
| 37 |
+
"loss": 1.2483,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.11,
|
| 42 |
+
"learning_rate": 4.814471243042672e-05,
|
| 43 |
+
"loss": 1.1685,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.13,
|
| 48 |
+
"learning_rate": 4.783549783549784e-05,
|
| 49 |
+
"loss": 1.0963,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.15,
|
| 54 |
+
"learning_rate": 4.752628324056896e-05,
|
| 55 |
+
"loss": 1.0671,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.17,
|
| 60 |
+
"learning_rate": 4.721706864564008e-05,
|
| 61 |
+
"loss": 1.0623,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.19,
|
| 66 |
+
"learning_rate": 4.69078540507112e-05,
|
| 67 |
+
"loss": 0.9754,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.2,
|
| 72 |
+
"learning_rate": 4.6598639455782315e-05,
|
| 73 |
+
"loss": 1.0497,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.22,
|
| 78 |
+
"learning_rate": 4.628942486085344e-05,
|
| 79 |
+
"loss": 1.0097,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.23,
|
| 84 |
+
"eval_avg_sts": 0.8361329859275041,
|
| 85 |
+
"eval_sickr_spearman": 0.8112100964923686,
|
| 86 |
+
"eval_stsb_spearman": 0.8610558753626395,
|
| 87 |
+
"step": 125
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.24,
|
| 91 |
+
"learning_rate": 4.5980210265924555e-05,
|
| 92 |
+
"loss": 0.981,
|
| 93 |
+
"step": 130
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.26,
|
| 97 |
+
"learning_rate": 4.567099567099568e-05,
|
| 98 |
+
"loss": 0.9648,
|
| 99 |
+
"step": 140
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.28,
|
| 103 |
+
"learning_rate": 4.5361781076066796e-05,
|
| 104 |
+
"loss": 0.9035,
|
| 105 |
+
"step": 150
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.3,
|
| 109 |
+
"learning_rate": 4.505256648113791e-05,
|
| 110 |
+
"loss": 0.9459,
|
| 111 |
+
"step": 160
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.32,
|
| 115 |
+
"learning_rate": 4.4743351886209036e-05,
|
| 116 |
+
"loss": 0.9296,
|
| 117 |
+
"step": 170
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.33,
|
| 121 |
+
"learning_rate": 4.4434137291280146e-05,
|
| 122 |
+
"loss": 0.9249,
|
| 123 |
+
"step": 180
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.35,
|
| 127 |
+
"learning_rate": 4.412492269635127e-05,
|
| 128 |
+
"loss": 0.8699,
|
| 129 |
+
"step": 190
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.37,
|
| 133 |
+
"learning_rate": 4.381570810142239e-05,
|
| 134 |
+
"loss": 0.88,
|
| 135 |
+
"step": 200
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.39,
|
| 139 |
+
"learning_rate": 4.3506493506493503e-05,
|
| 140 |
+
"loss": 0.8992,
|
| 141 |
+
"step": 210
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.41,
|
| 145 |
+
"learning_rate": 4.319727891156463e-05,
|
| 146 |
+
"loss": 0.893,
|
| 147 |
+
"step": 220
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.43,
|
| 151 |
+
"learning_rate": 4.2888064316635744e-05,
|
| 152 |
+
"loss": 0.8977,
|
| 153 |
+
"step": 230
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.45,
|
| 157 |
+
"learning_rate": 4.257884972170687e-05,
|
| 158 |
+
"loss": 0.8957,
|
| 159 |
+
"step": 240
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.46,
|
| 163 |
+
"learning_rate": 4.2269635126777984e-05,
|
| 164 |
+
"loss": 0.8969,
|
| 165 |
+
"step": 250
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.46,
|
| 169 |
+
"eval_avg_sts": 0.8389081420973826,
|
| 170 |
+
"eval_sickr_spearman": 0.8101423170762979,
|
| 171 |
+
"eval_stsb_spearman": 0.8676739671184673,
|
| 172 |
+
"step": 250
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.48,
|
| 176 |
+
"learning_rate": 4.19604205318491e-05,
|
| 177 |
+
"loss": 0.8434,
|
| 178 |
+
"step": 260
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.5,
|
| 182 |
+
"learning_rate": 4.1651205936920225e-05,
|
| 183 |
+
"loss": 0.8362,
|
| 184 |
+
"step": 270
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.52,
|
| 188 |
+
"learning_rate": 4.134199134199134e-05,
|
| 189 |
+
"loss": 0.8513,
|
| 190 |
+
"step": 280
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.54,
|
| 194 |
+
"learning_rate": 4.1032776747062465e-05,
|
| 195 |
+
"loss": 0.8384,
|
| 196 |
+
"step": 290
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.56,
|
| 200 |
+
"learning_rate": 4.072356215213358e-05,
|
| 201 |
+
"loss": 0.8653,
|
| 202 |
+
"step": 300
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.58,
|
| 206 |
+
"learning_rate": 4.04143475572047e-05,
|
| 207 |
+
"loss": 0.8558,
|
| 208 |
+
"step": 310
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.59,
|
| 212 |
+
"learning_rate": 4.010513296227582e-05,
|
| 213 |
+
"loss": 0.8241,
|
| 214 |
+
"step": 320
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.61,
|
| 218 |
+
"learning_rate": 3.979591836734694e-05,
|
| 219 |
+
"loss": 0.8324,
|
| 220 |
+
"step": 330
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.63,
|
| 224 |
+
"learning_rate": 3.948670377241806e-05,
|
| 225 |
+
"loss": 0.8396,
|
| 226 |
+
"step": 340
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.65,
|
| 230 |
+
"learning_rate": 3.917748917748918e-05,
|
| 231 |
+
"loss": 0.8361,
|
| 232 |
+
"step": 350
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.67,
|
| 236 |
+
"learning_rate": 3.88682745825603e-05,
|
| 237 |
+
"loss": 0.8737,
|
| 238 |
+
"step": 360
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.69,
|
| 242 |
+
"learning_rate": 3.855905998763142e-05,
|
| 243 |
+
"loss": 0.84,
|
| 244 |
+
"step": 370
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 0.7,
|
| 248 |
+
"eval_avg_sts": 0.8445458159728749,
|
| 249 |
+
"eval_sickr_spearman": 0.8180373333129849,
|
| 250 |
+
"eval_stsb_spearman": 0.8710542986327651,
|
| 251 |
+
"step": 375
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.71,
|
| 255 |
+
"learning_rate": 3.824984539270254e-05,
|
| 256 |
+
"loss": 0.8384,
|
| 257 |
+
"step": 380
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.72,
|
| 261 |
+
"learning_rate": 3.794063079777366e-05,
|
| 262 |
+
"loss": 0.8047,
|
| 263 |
+
"step": 390
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.74,
|
| 267 |
+
"learning_rate": 3.763141620284478e-05,
|
| 268 |
+
"loss": 0.8039,
|
| 269 |
+
"step": 400
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.76,
|
| 273 |
+
"learning_rate": 3.7322201607915894e-05,
|
| 274 |
+
"loss": 0.8474,
|
| 275 |
+
"step": 410
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.78,
|
| 279 |
+
"learning_rate": 3.701298701298702e-05,
|
| 280 |
+
"loss": 0.7912,
|
| 281 |
+
"step": 420
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.8,
|
| 285 |
+
"learning_rate": 3.6703772418058135e-05,
|
| 286 |
+
"loss": 0.8061,
|
| 287 |
+
"step": 430
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.82,
|
| 291 |
+
"learning_rate": 3.639455782312925e-05,
|
| 292 |
+
"loss": 0.8132,
|
| 293 |
+
"step": 440
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 0.83,
|
| 297 |
+
"learning_rate": 3.6085343228200375e-05,
|
| 298 |
+
"loss": 0.7815,
|
| 299 |
+
"step": 450
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.85,
|
| 303 |
+
"learning_rate": 3.577612863327149e-05,
|
| 304 |
+
"loss": 0.7839,
|
| 305 |
+
"step": 460
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.87,
|
| 309 |
+
"learning_rate": 3.5466914038342616e-05,
|
| 310 |
+
"loss": 0.7892,
|
| 311 |
+
"step": 470
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.89,
|
| 315 |
+
"learning_rate": 3.515769944341373e-05,
|
| 316 |
+
"loss": 0.7738,
|
| 317 |
+
"step": 480
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.91,
|
| 321 |
+
"learning_rate": 3.484848484848485e-05,
|
| 322 |
+
"loss": 0.7702,
|
| 323 |
+
"step": 490
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.93,
|
| 327 |
+
"learning_rate": 3.453927025355597e-05,
|
| 328 |
+
"loss": 0.7715,
|
| 329 |
+
"step": 500
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.93,
|
| 333 |
+
"eval_avg_sts": 0.8402489492013288,
|
| 334 |
+
"eval_sickr_spearman": 0.8102146999461652,
|
| 335 |
+
"eval_stsb_spearman": 0.8702831984564924,
|
| 336 |
+
"step": 500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.95,
|
| 340 |
+
"learning_rate": 3.423005565862709e-05,
|
| 341 |
+
"loss": 0.7891,
|
| 342 |
+
"step": 510
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 0.96,
|
| 346 |
+
"learning_rate": 3.392084106369821e-05,
|
| 347 |
+
"loss": 0.7646,
|
| 348 |
+
"step": 520
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 0.98,
|
| 352 |
+
"learning_rate": 3.3611626468769324e-05,
|
| 353 |
+
"loss": 0.7798,
|
| 354 |
+
"step": 530
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.0,
|
| 358 |
+
"learning_rate": 3.330241187384045e-05,
|
| 359 |
+
"loss": 0.7456,
|
| 360 |
+
"step": 540
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.02,
|
| 364 |
+
"learning_rate": 3.2993197278911564e-05,
|
| 365 |
+
"loss": 0.6767,
|
| 366 |
+
"step": 550
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.04,
|
| 370 |
+
"learning_rate": 3.268398268398268e-05,
|
| 371 |
+
"loss": 0.6897,
|
| 372 |
+
"step": 560
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.06,
|
| 376 |
+
"learning_rate": 3.2374768089053805e-05,
|
| 377 |
+
"loss": 0.6593,
|
| 378 |
+
"step": 570
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.08,
|
| 382 |
+
"learning_rate": 3.206555349412492e-05,
|
| 383 |
+
"loss": 0.697,
|
| 384 |
+
"step": 580
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.09,
|
| 388 |
+
"learning_rate": 3.1756338899196045e-05,
|
| 389 |
+
"loss": 0.6816,
|
| 390 |
+
"step": 590
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 1.11,
|
| 394 |
+
"learning_rate": 3.144712430426716e-05,
|
| 395 |
+
"loss": 0.6675,
|
| 396 |
+
"step": 600
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.13,
|
| 400 |
+
"learning_rate": 3.113790970933828e-05,
|
| 401 |
+
"loss": 0.6661,
|
| 402 |
+
"step": 610
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.15,
|
| 406 |
+
"learning_rate": 3.08286951144094e-05,
|
| 407 |
+
"loss": 0.6719,
|
| 408 |
+
"step": 620
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.16,
|
| 412 |
+
"eval_avg_sts": 0.8420328093315328,
|
| 413 |
+
"eval_sickr_spearman": 0.8171519279890747,
|
| 414 |
+
"eval_stsb_spearman": 0.8669136906739908,
|
| 415 |
+
"step": 625
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.17,
|
| 419 |
+
"learning_rate": 3.051948051948052e-05,
|
| 420 |
+
"loss": 0.7076,
|
| 421 |
+
"step": 630
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 1.19,
|
| 425 |
+
"learning_rate": 3.021026592455164e-05,
|
| 426 |
+
"loss": 0.6515,
|
| 427 |
+
"step": 640
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.21,
|
| 431 |
+
"learning_rate": 2.990105132962276e-05,
|
| 432 |
+
"loss": 0.6893,
|
| 433 |
+
"step": 650
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 1.22,
|
| 437 |
+
"learning_rate": 2.959183673469388e-05,
|
| 438 |
+
"loss": 0.6714,
|
| 439 |
+
"step": 660
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 1.24,
|
| 443 |
+
"learning_rate": 2.9282622139764997e-05,
|
| 444 |
+
"loss": 0.7052,
|
| 445 |
+
"step": 670
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 1.26,
|
| 449 |
+
"learning_rate": 2.8973407544836117e-05,
|
| 450 |
+
"loss": 0.6671,
|
| 451 |
+
"step": 680
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.28,
|
| 455 |
+
"learning_rate": 2.8664192949907237e-05,
|
| 456 |
+
"loss": 0.6735,
|
| 457 |
+
"step": 690
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.3,
|
| 461 |
+
"learning_rate": 2.8354978354978357e-05,
|
| 462 |
+
"loss": 0.6778,
|
| 463 |
+
"step": 700
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 1.32,
|
| 467 |
+
"learning_rate": 2.8045763760049478e-05,
|
| 468 |
+
"loss": 0.6832,
|
| 469 |
+
"step": 710
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 1.34,
|
| 473 |
+
"learning_rate": 2.7736549165120594e-05,
|
| 474 |
+
"loss": 0.6587,
|
| 475 |
+
"step": 720
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 1.35,
|
| 479 |
+
"learning_rate": 2.7427334570191715e-05,
|
| 480 |
+
"loss": 0.6567,
|
| 481 |
+
"step": 730
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 1.37,
|
| 485 |
+
"learning_rate": 2.7118119975262835e-05,
|
| 486 |
+
"loss": 0.648,
|
| 487 |
+
"step": 740
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.39,
|
| 491 |
+
"learning_rate": 2.6808905380333955e-05,
|
| 492 |
+
"loss": 0.6488,
|
| 493 |
+
"step": 750
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.39,
|
| 497 |
+
"eval_avg_sts": 0.8417559206280472,
|
| 498 |
+
"eval_sickr_spearman": 0.8131766819096506,
|
| 499 |
+
"eval_stsb_spearman": 0.8703351593464438,
|
| 500 |
+
"step": 750
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.41,
|
| 504 |
+
"learning_rate": 2.6499690785405072e-05,
|
| 505 |
+
"loss": 0.6445,
|
| 506 |
+
"step": 760
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.43,
|
| 510 |
+
"learning_rate": 2.6190476190476192e-05,
|
| 511 |
+
"loss": 0.6763,
|
| 512 |
+
"step": 770
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.45,
|
| 516 |
+
"learning_rate": 2.5881261595547312e-05,
|
| 517 |
+
"loss": 0.6585,
|
| 518 |
+
"step": 780
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.47,
|
| 522 |
+
"learning_rate": 2.5572047000618433e-05,
|
| 523 |
+
"loss": 0.6492,
|
| 524 |
+
"step": 790
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 1.48,
|
| 528 |
+
"learning_rate": 2.5262832405689553e-05,
|
| 529 |
+
"loss": 0.6634,
|
| 530 |
+
"step": 800
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 1.5,
|
| 534 |
+
"learning_rate": 2.495361781076067e-05,
|
| 535 |
+
"loss": 0.6757,
|
| 536 |
+
"step": 810
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.52,
|
| 540 |
+
"learning_rate": 2.4644403215831787e-05,
|
| 541 |
+
"loss": 0.6805,
|
| 542 |
+
"step": 820
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.54,
|
| 546 |
+
"learning_rate": 2.4335188620902907e-05,
|
| 547 |
+
"loss": 0.6639,
|
| 548 |
+
"step": 830
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.56,
|
| 552 |
+
"learning_rate": 2.4025974025974027e-05,
|
| 553 |
+
"loss": 0.6576,
|
| 554 |
+
"step": 840
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 1.58,
|
| 558 |
+
"learning_rate": 2.3716759431045147e-05,
|
| 559 |
+
"loss": 0.6849,
|
| 560 |
+
"step": 850
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 1.6,
|
| 564 |
+
"learning_rate": 2.3407544836116267e-05,
|
| 565 |
+
"loss": 0.7029,
|
| 566 |
+
"step": 860
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 1.61,
|
| 570 |
+
"learning_rate": 2.3098330241187384e-05,
|
| 571 |
+
"loss": 0.6384,
|
| 572 |
+
"step": 870
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.62,
|
| 576 |
+
"eval_avg_sts": 0.8415267776775321,
|
| 577 |
+
"eval_sickr_spearman": 0.8138372056166275,
|
| 578 |
+
"eval_stsb_spearman": 0.8692163497384369,
|
| 579 |
+
"step": 875
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 1.63,
|
| 583 |
+
"learning_rate": 2.2789115646258505e-05,
|
| 584 |
+
"loss": 0.6228,
|
| 585 |
+
"step": 880
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 1.65,
|
| 589 |
+
"learning_rate": 2.2479901051329625e-05,
|
| 590 |
+
"loss": 0.6844,
|
| 591 |
+
"step": 890
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.67,
|
| 595 |
+
"learning_rate": 2.2170686456400745e-05,
|
| 596 |
+
"loss": 0.646,
|
| 597 |
+
"step": 900
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.69,
|
| 601 |
+
"learning_rate": 2.1861471861471862e-05,
|
| 602 |
+
"loss": 0.6328,
|
| 603 |
+
"step": 910
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 1.71,
|
| 607 |
+
"learning_rate": 2.1552257266542982e-05,
|
| 608 |
+
"loss": 0.6774,
|
| 609 |
+
"step": 920
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 1.73,
|
| 613 |
+
"learning_rate": 2.1243042671614102e-05,
|
| 614 |
+
"loss": 0.6613,
|
| 615 |
+
"step": 930
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 1.74,
|
| 619 |
+
"learning_rate": 2.0933828076685223e-05,
|
| 620 |
+
"loss": 0.6602,
|
| 621 |
+
"step": 940
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 1.76,
|
| 625 |
+
"learning_rate": 2.062461348175634e-05,
|
| 626 |
+
"loss": 0.6581,
|
| 627 |
+
"step": 950
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 1.78,
|
| 631 |
+
"learning_rate": 2.031539888682746e-05,
|
| 632 |
+
"loss": 0.6729,
|
| 633 |
+
"step": 960
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.8,
|
| 637 |
+
"learning_rate": 2.0006184291898576e-05,
|
| 638 |
+
"loss": 0.6428,
|
| 639 |
+
"step": 970
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.82,
|
| 643 |
+
"learning_rate": 1.9696969696969697e-05,
|
| 644 |
+
"loss": 0.6552,
|
| 645 |
+
"step": 980
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 1.84,
|
| 649 |
+
"learning_rate": 1.9387755102040817e-05,
|
| 650 |
+
"loss": 0.6501,
|
| 651 |
+
"step": 990
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 1.86,
|
| 655 |
+
"learning_rate": 1.9078540507111937e-05,
|
| 656 |
+
"loss": 0.6839,
|
| 657 |
+
"step": 1000
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 1.86,
|
| 661 |
+
"eval_avg_sts": 0.8442180854633452,
|
| 662 |
+
"eval_sickr_spearman": 0.814642927343252,
|
| 663 |
+
"eval_stsb_spearman": 0.8737932435834382,
|
| 664 |
+
"step": 1000
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 1.87,
|
| 668 |
+
"learning_rate": 1.8769325912183054e-05,
|
| 669 |
+
"loss": 0.646,
|
| 670 |
+
"step": 1010
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 1.89,
|
| 674 |
+
"learning_rate": 1.8460111317254174e-05,
|
| 675 |
+
"loss": 0.6217,
|
| 676 |
+
"step": 1020
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.91,
|
| 680 |
+
"learning_rate": 1.8150896722325294e-05,
|
| 681 |
+
"loss": 0.6278,
|
| 682 |
+
"step": 1030
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.93,
|
| 686 |
+
"learning_rate": 1.7841682127396415e-05,
|
| 687 |
+
"loss": 0.6456,
|
| 688 |
+
"step": 1040
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.95,
|
| 692 |
+
"learning_rate": 1.7532467532467535e-05,
|
| 693 |
+
"loss": 0.6459,
|
| 694 |
+
"step": 1050
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 1.97,
|
| 698 |
+
"learning_rate": 1.7223252937538652e-05,
|
| 699 |
+
"loss": 0.6223,
|
| 700 |
+
"step": 1060
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 1.99,
|
| 704 |
+
"learning_rate": 1.6914038342609772e-05,
|
| 705 |
+
"loss": 0.6357,
|
| 706 |
+
"step": 1070
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 2.0,
|
| 710 |
+
"learning_rate": 1.6604823747680892e-05,
|
| 711 |
+
"loss": 0.6018,
|
| 712 |
+
"step": 1080
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 2.02,
|
| 716 |
+
"learning_rate": 1.6295609152752012e-05,
|
| 717 |
+
"loss": 0.5888,
|
| 718 |
+
"step": 1090
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 2.04,
|
| 722 |
+
"learning_rate": 1.5986394557823133e-05,
|
| 723 |
+
"loss": 0.5913,
|
| 724 |
+
"step": 1100
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 2.06,
|
| 728 |
+
"learning_rate": 1.567717996289425e-05,
|
| 729 |
+
"loss": 0.5581,
|
| 730 |
+
"step": 1110
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.08,
|
| 734 |
+
"learning_rate": 1.5367965367965366e-05,
|
| 735 |
+
"loss": 0.5899,
|
| 736 |
+
"step": 1120
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 2.09,
|
| 740 |
+
"eval_avg_sts": 0.8410988161028077,
|
| 741 |
+
"eval_sickr_spearman": 0.8122731208293968,
|
| 742 |
+
"eval_stsb_spearman": 0.8699245113762184,
|
| 743 |
+
"step": 1125
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 2.1,
|
| 747 |
+
"learning_rate": 1.5058750773036487e-05,
|
| 748 |
+
"loss": 0.5456,
|
| 749 |
+
"step": 1130
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 2.12,
|
| 753 |
+
"learning_rate": 1.4749536178107607e-05,
|
| 754 |
+
"loss": 0.6023,
|
| 755 |
+
"step": 1140
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 2.13,
|
| 759 |
+
"learning_rate": 1.4440321583178725e-05,
|
| 760 |
+
"loss": 0.6004,
|
| 761 |
+
"step": 1150
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 2.15,
|
| 765 |
+
"learning_rate": 1.4131106988249846e-05,
|
| 766 |
+
"loss": 0.588,
|
| 767 |
+
"step": 1160
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 2.17,
|
| 771 |
+
"learning_rate": 1.3821892393320964e-05,
|
| 772 |
+
"loss": 0.5716,
|
| 773 |
+
"step": 1170
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 2.19,
|
| 777 |
+
"learning_rate": 1.3512677798392084e-05,
|
| 778 |
+
"loss": 0.5813,
|
| 779 |
+
"step": 1180
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.21,
|
| 783 |
+
"learning_rate": 1.3203463203463205e-05,
|
| 784 |
+
"loss": 0.5615,
|
| 785 |
+
"step": 1190
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 2.23,
|
| 789 |
+
"learning_rate": 1.2894248608534323e-05,
|
| 790 |
+
"loss": 0.5874,
|
| 791 |
+
"step": 1200
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 2.24,
|
| 795 |
+
"learning_rate": 1.2585034013605443e-05,
|
| 796 |
+
"loss": 0.5983,
|
| 797 |
+
"step": 1210
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 2.26,
|
| 801 |
+
"learning_rate": 1.2275819418676562e-05,
|
| 802 |
+
"loss": 0.5922,
|
| 803 |
+
"step": 1220
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 2.28,
|
| 807 |
+
"learning_rate": 1.1966604823747682e-05,
|
| 808 |
+
"loss": 0.5772,
|
| 809 |
+
"step": 1230
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 2.3,
|
| 813 |
+
"learning_rate": 1.16573902288188e-05,
|
| 814 |
+
"loss": 0.5772,
|
| 815 |
+
"step": 1240
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 2.32,
|
| 819 |
+
"learning_rate": 1.1348175633889919e-05,
|
| 820 |
+
"loss": 0.5828,
|
| 821 |
+
"step": 1250
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.32,
|
| 825 |
+
"eval_avg_sts": 0.8417651170306701,
|
| 826 |
+
"eval_sickr_spearman": 0.8143169402577914,
|
| 827 |
+
"eval_stsb_spearman": 0.869213293803549,
|
| 828 |
+
"step": 1250
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 2.34,
|
| 832 |
+
"learning_rate": 1.103896103896104e-05,
|
| 833 |
+
"loss": 0.5919,
|
| 834 |
+
"step": 1260
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 2.36,
|
| 838 |
+
"learning_rate": 1.0729746444032158e-05,
|
| 839 |
+
"loss": 0.5587,
|
| 840 |
+
"step": 1270
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 2.37,
|
| 844 |
+
"learning_rate": 1.0420531849103278e-05,
|
| 845 |
+
"loss": 0.5779,
|
| 846 |
+
"step": 1280
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 2.39,
|
| 850 |
+
"learning_rate": 1.0111317254174398e-05,
|
| 851 |
+
"loss": 0.5422,
|
| 852 |
+
"step": 1290
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 2.41,
|
| 856 |
+
"learning_rate": 9.802102659245517e-06,
|
| 857 |
+
"loss": 0.5915,
|
| 858 |
+
"step": 1300
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 2.43,
|
| 862 |
+
"learning_rate": 9.492888064316637e-06,
|
| 863 |
+
"loss": 0.5794,
|
| 864 |
+
"step": 1310
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 2.45,
|
| 868 |
+
"learning_rate": 9.183673469387756e-06,
|
| 869 |
+
"loss": 0.5832,
|
| 870 |
+
"step": 1320
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 2.47,
|
| 874 |
+
"learning_rate": 8.874458874458876e-06,
|
| 875 |
+
"loss": 0.5644,
|
| 876 |
+
"step": 1330
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 2.49,
|
| 880 |
+
"learning_rate": 8.565244279529994e-06,
|
| 881 |
+
"loss": 0.5704,
|
| 882 |
+
"step": 1340
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 2.5,
|
| 886 |
+
"learning_rate": 8.256029684601113e-06,
|
| 887 |
+
"loss": 0.5692,
|
| 888 |
+
"step": 1350
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 2.52,
|
| 892 |
+
"learning_rate": 7.946815089672233e-06,
|
| 893 |
+
"loss": 0.5794,
|
| 894 |
+
"step": 1360
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 2.54,
|
| 898 |
+
"learning_rate": 7.637600494743352e-06,
|
| 899 |
+
"loss": 0.5793,
|
| 900 |
+
"step": 1370
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 2.55,
|
| 904 |
+
"eval_avg_sts": 0.8438379193565275,
|
| 905 |
+
"eval_sickr_spearman": 0.8155493222584906,
|
| 906 |
+
"eval_stsb_spearman": 0.8721265164545645,
|
| 907 |
+
"step": 1375
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 2.56,
|
| 911 |
+
"learning_rate": 7.328385899814472e-06,
|
| 912 |
+
"loss": 0.57,
|
| 913 |
+
"step": 1380
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 2.58,
|
| 917 |
+
"learning_rate": 7.019171304885591e-06,
|
| 918 |
+
"loss": 0.5759,
|
| 919 |
+
"step": 1390
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.6,
|
| 923 |
+
"learning_rate": 6.709956709956711e-06,
|
| 924 |
+
"loss": 0.5518,
|
| 925 |
+
"step": 1400
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 2.62,
|
| 929 |
+
"learning_rate": 6.40074211502783e-06,
|
| 930 |
+
"loss": 0.5608,
|
| 931 |
+
"step": 1410
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 2.63,
|
| 935 |
+
"learning_rate": 6.091527520098949e-06,
|
| 936 |
+
"loss": 0.5544,
|
| 937 |
+
"step": 1420
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 2.65,
|
| 941 |
+
"learning_rate": 5.782312925170069e-06,
|
| 942 |
+
"loss": 0.5701,
|
| 943 |
+
"step": 1430
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 2.67,
|
| 947 |
+
"learning_rate": 5.473098330241188e-06,
|
| 948 |
+
"loss": 0.5886,
|
| 949 |
+
"step": 1440
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 2.69,
|
| 953 |
+
"learning_rate": 5.163883735312307e-06,
|
| 954 |
+
"loss": 0.5781,
|
| 955 |
+
"step": 1450
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 2.71,
|
| 959 |
+
"learning_rate": 4.854669140383426e-06,
|
| 960 |
+
"loss": 0.5767,
|
| 961 |
+
"step": 1460
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.73,
|
| 965 |
+
"learning_rate": 4.5454545454545455e-06,
|
| 966 |
+
"loss": 0.5859,
|
| 967 |
+
"step": 1470
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 2.75,
|
| 971 |
+
"learning_rate": 4.236239950525665e-06,
|
| 972 |
+
"loss": 0.5906,
|
| 973 |
+
"step": 1480
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 2.76,
|
| 977 |
+
"learning_rate": 3.927025355596784e-06,
|
| 978 |
+
"loss": 0.5621,
|
| 979 |
+
"step": 1490
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 2.78,
|
| 983 |
+
"learning_rate": 3.6178107606679037e-06,
|
| 984 |
+
"loss": 0.5505,
|
| 985 |
+
"step": 1500
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 2.78,
|
| 989 |
+
"eval_avg_sts": 0.8425113202508864,
|
| 990 |
+
"eval_sickr_spearman": 0.813630912035951,
|
| 991 |
+
"eval_stsb_spearman": 0.8713917284658218,
|
| 992 |
+
"step": 1500
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 2.8,
|
| 996 |
+
"learning_rate": 3.308596165739023e-06,
|
| 997 |
+
"loss": 0.5843,
|
| 998 |
+
"step": 1510
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 2.82,
|
| 1002 |
+
"learning_rate": 2.9993815708101424e-06,
|
| 1003 |
+
"loss": 0.5518,
|
| 1004 |
+
"step": 1520
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 2.84,
|
| 1008 |
+
"learning_rate": 2.690166975881262e-06,
|
| 1009 |
+
"loss": 0.5837,
|
| 1010 |
+
"step": 1530
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.86,
|
| 1014 |
+
"learning_rate": 2.3809523809523808e-06,
|
| 1015 |
+
"loss": 0.5736,
|
| 1016 |
+
"step": 1540
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 2.88,
|
| 1020 |
+
"learning_rate": 2.0717377860235006e-06,
|
| 1021 |
+
"loss": 0.5655,
|
| 1022 |
+
"step": 1550
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 2.89,
|
| 1026 |
+
"learning_rate": 1.7625231910946195e-06,
|
| 1027 |
+
"loss": 0.5563,
|
| 1028 |
+
"step": 1560
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 2.91,
|
| 1032 |
+
"learning_rate": 1.4533085961657391e-06,
|
| 1033 |
+
"loss": 0.5646,
|
| 1034 |
+
"step": 1570
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 2.93,
|
| 1038 |
+
"learning_rate": 1.1440940012368585e-06,
|
| 1039 |
+
"loss": 0.5493,
|
| 1040 |
+
"step": 1580
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 2.95,
|
| 1044 |
+
"learning_rate": 8.348794063079778e-07,
|
| 1045 |
+
"loss": 0.5594,
|
| 1046 |
+
"step": 1590
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 2.97,
|
| 1050 |
+
"learning_rate": 5.25664811379097e-07,
|
| 1051 |
+
"loss": 0.5628,
|
| 1052 |
+
"step": 1600
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.99,
|
| 1056 |
+
"learning_rate": 2.1645021645021646e-07,
|
| 1057 |
+
"loss": 0.59,
|
| 1058 |
+
"step": 1610
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 3.0,
|
| 1062 |
+
"step": 1617,
|
| 1063 |
+
"train_runtime": 1218.5135,
|
| 1064 |
+
"train_samples_per_second": 1.327
|
| 1065 |
+
}
|
| 1066 |
+
],
|
| 1067 |
+
"max_steps": 1617,
|
| 1068 |
+
"num_train_epochs": 3,
|
| 1069 |
+
"total_flos": 59643369959989248,
|
| 1070 |
+
"trial_name": null,
|
| 1071 |
+
"trial_params": null
|
| 1072 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e3eeaaa83b4d51c23185cd2772c610b7157e9027e3f5a450ec347affcbaaebf0
|
| 3 |
+
size 2287
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|