hxl520
commited on
Commit
·
8fa1142
1
Parent(s):
3f3210c
Update from test-mlm-wwm2
Browse files- config.json +33 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- train_results.txt +5 -0
- trainer_state.json +1111 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "chinese-roberta-wwm-ext",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"directionality": "bidi",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"gradient_checkpointing": false,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_norm_eps": 1e-12,
|
| 17 |
+
"max_position_embeddings": 512,
|
| 18 |
+
"model_type": "bert",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"output_past": true,
|
| 22 |
+
"pad_token_id": 0,
|
| 23 |
+
"pooler_fc_size": 768,
|
| 24 |
+
"pooler_num_attention_heads": 12,
|
| 25 |
+
"pooler_num_fc_layers": 3,
|
| 26 |
+
"pooler_size_per_head": 128,
|
| 27 |
+
"pooler_type": "first_token_transform",
|
| 28 |
+
"position_embedding_type": "absolute",
|
| 29 |
+
"transformers_version": "4.7.0",
|
| 30 |
+
"type_vocab_size": 2,
|
| 31 |
+
"use_cache": true,
|
| 32 |
+
"vocab_size": 21128
|
| 33 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1991fa9f47b802ebc0d828d1605a993c63df40e086d82dd9d25eb975425cdcb9
|
| 3 |
+
size 409246059
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": "/root/.cache/huggingface/transformers/d521373fc7ac35f63d56cf303de74a202403dcf1aaa792cd01f653694be59563.dd8bd9bfd3664b530ea4e645105f557769387b3da9f79bdb55ed556bdd80611d", "name_or_path": "hy-tmp/test-mlm-wwm/checkpoint-26000"}
|
train_results.txt
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
epoch = 3.0
|
| 2 |
+
train_loss = 1.3765672623247966
|
| 3 |
+
train_runtime = 7341.0199
|
| 4 |
+
train_samples_per_second = 99.099
|
| 5 |
+
train_steps_per_second = 12.388
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1111 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"global_step": 90939,
|
| 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": 4.972788227098572e-05,
|
| 13 |
+
"loss": 2.3513,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.03,
|
| 18 |
+
"learning_rate": 4.945576454197144e-05,
|
| 19 |
+
"loss": 2.3368,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.05,
|
| 24 |
+
"learning_rate": 4.9183646812957165e-05,
|
| 25 |
+
"loss": 2.3554,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.07,
|
| 30 |
+
"learning_rate": 4.891152908394288e-05,
|
| 31 |
+
"loss": 2.4827,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.08,
|
| 36 |
+
"learning_rate": 4.86394113549286e-05,
|
| 37 |
+
"loss": 2.5032,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.1,
|
| 42 |
+
"learning_rate": 4.836729362591432e-05,
|
| 43 |
+
"loss": 2.4807,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.11,
|
| 48 |
+
"learning_rate": 4.809517589690004e-05,
|
| 49 |
+
"loss": 2.4771,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.13,
|
| 54 |
+
"learning_rate": 4.782305816788576e-05,
|
| 55 |
+
"loss": 2.432,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.15,
|
| 60 |
+
"learning_rate": 4.7550940438871475e-05,
|
| 61 |
+
"loss": 2.4253,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.16,
|
| 66 |
+
"learning_rate": 4.7278822709857194e-05,
|
| 67 |
+
"loss": 2.396,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.18,
|
| 72 |
+
"learning_rate": 4.700670498084291e-05,
|
| 73 |
+
"loss": 2.3774,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.2,
|
| 78 |
+
"learning_rate": 4.673458725182863e-05,
|
| 79 |
+
"loss": 2.3601,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.21,
|
| 84 |
+
"learning_rate": 4.646246952281435e-05,
|
| 85 |
+
"loss": 2.3444,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.23,
|
| 90 |
+
"learning_rate": 4.6190351793800074e-05,
|
| 91 |
+
"loss": 2.3522,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.24,
|
| 96 |
+
"learning_rate": 4.591823406478579e-05,
|
| 97 |
+
"loss": 2.4588,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.26,
|
| 102 |
+
"learning_rate": 4.564611633577151e-05,
|
| 103 |
+
"loss": 2.3588,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.28,
|
| 108 |
+
"learning_rate": 4.537399860675723e-05,
|
| 109 |
+
"loss": 2.3733,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.29,
|
| 114 |
+
"learning_rate": 4.510188087774295e-05,
|
| 115 |
+
"loss": 2.3612,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.31,
|
| 120 |
+
"learning_rate": 4.482976314872867e-05,
|
| 121 |
+
"loss": 2.413,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.33,
|
| 126 |
+
"learning_rate": 4.455764541971439e-05,
|
| 127 |
+
"loss": 2.3861,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.34,
|
| 132 |
+
"learning_rate": 4.428552769070011e-05,
|
| 133 |
+
"loss": 2.3726,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.36,
|
| 138 |
+
"learning_rate": 4.401340996168583e-05,
|
| 139 |
+
"loss": 2.328,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.38,
|
| 144 |
+
"learning_rate": 4.374129223267154e-05,
|
| 145 |
+
"loss": 2.3552,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.39,
|
| 150 |
+
"learning_rate": 4.346917450365726e-05,
|
| 151 |
+
"loss": 2.4816,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.41,
|
| 156 |
+
"learning_rate": 4.3197056774642984e-05,
|
| 157 |
+
"loss": 2.3795,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.42,
|
| 162 |
+
"learning_rate": 4.29249390456287e-05,
|
| 163 |
+
"loss": 2.3067,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.44,
|
| 168 |
+
"learning_rate": 4.265282131661442e-05,
|
| 169 |
+
"loss": 2.3856,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.46,
|
| 174 |
+
"learning_rate": 4.238070358760014e-05,
|
| 175 |
+
"loss": 2.2952,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.47,
|
| 180 |
+
"learning_rate": 4.210858585858586e-05,
|
| 181 |
+
"loss": 2.2996,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.49,
|
| 186 |
+
"learning_rate": 4.183646812957158e-05,
|
| 187 |
+
"loss": 2.3925,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.51,
|
| 192 |
+
"learning_rate": 4.15643504005573e-05,
|
| 193 |
+
"loss": 2.3015,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.52,
|
| 198 |
+
"learning_rate": 4.129223267154302e-05,
|
| 199 |
+
"loss": 2.4006,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.54,
|
| 204 |
+
"learning_rate": 4.102011494252874e-05,
|
| 205 |
+
"loss": 2.3463,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.56,
|
| 210 |
+
"learning_rate": 4.0747997213514457e-05,
|
| 211 |
+
"loss": 2.3236,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.57,
|
| 216 |
+
"learning_rate": 4.047587948450018e-05,
|
| 217 |
+
"loss": 2.3262,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.59,
|
| 222 |
+
"learning_rate": 4.02037617554859e-05,
|
| 223 |
+
"loss": 2.3238,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.6,
|
| 228 |
+
"learning_rate": 3.993164402647161e-05,
|
| 229 |
+
"loss": 2.185,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.62,
|
| 234 |
+
"learning_rate": 3.965952629745733e-05,
|
| 235 |
+
"loss": 2.2306,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.64,
|
| 240 |
+
"learning_rate": 3.938740856844305e-05,
|
| 241 |
+
"loss": 2.2768,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.65,
|
| 246 |
+
"learning_rate": 3.911529083942877e-05,
|
| 247 |
+
"loss": 2.2603,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.67,
|
| 252 |
+
"learning_rate": 3.884317311041449e-05,
|
| 253 |
+
"loss": 2.3123,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.69,
|
| 258 |
+
"learning_rate": 3.857105538140021e-05,
|
| 259 |
+
"loss": 2.2436,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.7,
|
| 264 |
+
"learning_rate": 3.829893765238593e-05,
|
| 265 |
+
"loss": 2.3416,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.72,
|
| 270 |
+
"learning_rate": 3.802681992337165e-05,
|
| 271 |
+
"loss": 2.2964,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.73,
|
| 276 |
+
"learning_rate": 3.7754702194357366e-05,
|
| 277 |
+
"loss": 2.4278,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.75,
|
| 282 |
+
"learning_rate": 3.748258446534309e-05,
|
| 283 |
+
"loss": 2.3682,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.77,
|
| 288 |
+
"learning_rate": 3.721046673632881e-05,
|
| 289 |
+
"loss": 2.3585,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.78,
|
| 294 |
+
"learning_rate": 3.693834900731453e-05,
|
| 295 |
+
"loss": 2.2861,
|
| 296 |
+
"step": 24000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.8,
|
| 300 |
+
"learning_rate": 3.666623127830025e-05,
|
| 301 |
+
"loss": 2.2447,
|
| 302 |
+
"step": 24500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.82,
|
| 306 |
+
"learning_rate": 3.6394113549285965e-05,
|
| 307 |
+
"loss": 2.3573,
|
| 308 |
+
"step": 25000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.83,
|
| 312 |
+
"learning_rate": 3.6121995820271684e-05,
|
| 313 |
+
"loss": 2.2819,
|
| 314 |
+
"step": 25500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.85,
|
| 318 |
+
"learning_rate": 3.58498780912574e-05,
|
| 319 |
+
"loss": 2.3119,
|
| 320 |
+
"step": 26000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.87,
|
| 324 |
+
"learning_rate": 3.542979359790629e-05,
|
| 325 |
+
"loss": 2.2534,
|
| 326 |
+
"step": 26500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.89,
|
| 330 |
+
"learning_rate": 3.5154884043149804e-05,
|
| 331 |
+
"loss": 2.2352,
|
| 332 |
+
"step": 27000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.91,
|
| 336 |
+
"learning_rate": 3.4879974488393316e-05,
|
| 337 |
+
"loss": 2.2077,
|
| 338 |
+
"step": 27500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.92,
|
| 342 |
+
"learning_rate": 3.4605064933636835e-05,
|
| 343 |
+
"loss": 2.2063,
|
| 344 |
+
"step": 28000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.94,
|
| 348 |
+
"learning_rate": 3.4330155378880354e-05,
|
| 349 |
+
"loss": 2.2603,
|
| 350 |
+
"step": 28500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.96,
|
| 354 |
+
"learning_rate": 3.4055245824123866e-05,
|
| 355 |
+
"loss": 2.2869,
|
| 356 |
+
"step": 29000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.97,
|
| 360 |
+
"learning_rate": 3.378033626936738e-05,
|
| 361 |
+
"loss": 2.2388,
|
| 362 |
+
"step": 29500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.99,
|
| 366 |
+
"learning_rate": 3.35054267146109e-05,
|
| 367 |
+
"loss": 2.2192,
|
| 368 |
+
"step": 30000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.01,
|
| 372 |
+
"learning_rate": 3.323051715985441e-05,
|
| 373 |
+
"loss": 2.2518,
|
| 374 |
+
"step": 30500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.02,
|
| 378 |
+
"learning_rate": 3.295560760509793e-05,
|
| 379 |
+
"loss": 2.2317,
|
| 380 |
+
"step": 31000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.04,
|
| 384 |
+
"learning_rate": 3.268069805034144e-05,
|
| 385 |
+
"loss": 2.1664,
|
| 386 |
+
"step": 31500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 1.06,
|
| 390 |
+
"learning_rate": 3.240578849558495e-05,
|
| 391 |
+
"loss": 2.2451,
|
| 392 |
+
"step": 32000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 1.07,
|
| 396 |
+
"learning_rate": 3.213087894082847e-05,
|
| 397 |
+
"loss": 2.1932,
|
| 398 |
+
"step": 32500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 1.09,
|
| 402 |
+
"learning_rate": 3.185596938607198e-05,
|
| 403 |
+
"loss": 2.1175,
|
| 404 |
+
"step": 33000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.11,
|
| 408 |
+
"learning_rate": 3.1581059831315495e-05,
|
| 409 |
+
"loss": 2.1072,
|
| 410 |
+
"step": 33500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.12,
|
| 414 |
+
"learning_rate": 3.1306150276559014e-05,
|
| 415 |
+
"loss": 2.2148,
|
| 416 |
+
"step": 34000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.14,
|
| 420 |
+
"learning_rate": 3.103124072180253e-05,
|
| 421 |
+
"loss": 2.1475,
|
| 422 |
+
"step": 34500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.15,
|
| 426 |
+
"learning_rate": 3.0756331167046045e-05,
|
| 427 |
+
"loss": 2.133,
|
| 428 |
+
"step": 35000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 1.17,
|
| 432 |
+
"learning_rate": 3.0481421612289557e-05,
|
| 433 |
+
"loss": 2.0703,
|
| 434 |
+
"step": 35500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.19,
|
| 438 |
+
"learning_rate": 3.0206512057533072e-05,
|
| 439 |
+
"loss": 2.1113,
|
| 440 |
+
"step": 36000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.2,
|
| 444 |
+
"learning_rate": 2.993160250277659e-05,
|
| 445 |
+
"loss": 2.0876,
|
| 446 |
+
"step": 36500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.22,
|
| 450 |
+
"learning_rate": 2.9656692948020103e-05,
|
| 451 |
+
"loss": 2.1706,
|
| 452 |
+
"step": 37000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.24,
|
| 456 |
+
"learning_rate": 2.938178339326362e-05,
|
| 457 |
+
"loss": 2.1105,
|
| 458 |
+
"step": 37500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.25,
|
| 462 |
+
"learning_rate": 2.910687383850713e-05,
|
| 463 |
+
"loss": 2.105,
|
| 464 |
+
"step": 38000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.27,
|
| 468 |
+
"learning_rate": 2.8831964283750646e-05,
|
| 469 |
+
"loss": 2.144,
|
| 470 |
+
"step": 38500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.29,
|
| 474 |
+
"learning_rate": 2.8557054728994165e-05,
|
| 475 |
+
"loss": 2.1656,
|
| 476 |
+
"step": 39000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.3,
|
| 480 |
+
"learning_rate": 2.8282145174237677e-05,
|
| 481 |
+
"loss": 2.073,
|
| 482 |
+
"step": 39500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.32,
|
| 486 |
+
"learning_rate": 2.8007235619481192e-05,
|
| 487 |
+
"loss": 2.1128,
|
| 488 |
+
"step": 40000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.34,
|
| 492 |
+
"learning_rate": 2.7732326064724708e-05,
|
| 493 |
+
"loss": 2.028,
|
| 494 |
+
"step": 40500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.35,
|
| 498 |
+
"learning_rate": 2.745741650996822e-05,
|
| 499 |
+
"loss": 2.1078,
|
| 500 |
+
"step": 41000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.37,
|
| 504 |
+
"learning_rate": 2.7182506955211735e-05,
|
| 505 |
+
"loss": 2.0751,
|
| 506 |
+
"step": 41500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.39,
|
| 510 |
+
"learning_rate": 2.6907597400455254e-05,
|
| 511 |
+
"loss": 2.0514,
|
| 512 |
+
"step": 42000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.4,
|
| 516 |
+
"learning_rate": 2.6632687845698766e-05,
|
| 517 |
+
"loss": 2.1142,
|
| 518 |
+
"step": 42500
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.42,
|
| 522 |
+
"learning_rate": 2.6357778290942282e-05,
|
| 523 |
+
"loss": 2.0525,
|
| 524 |
+
"step": 43000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 1.44,
|
| 528 |
+
"learning_rate": 2.6082868736185794e-05,
|
| 529 |
+
"loss": 2.1227,
|
| 530 |
+
"step": 43500
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 1.45,
|
| 534 |
+
"learning_rate": 2.580795918142931e-05,
|
| 535 |
+
"loss": 2.0927,
|
| 536 |
+
"step": 44000
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.47,
|
| 540 |
+
"learning_rate": 2.553304962667283e-05,
|
| 541 |
+
"loss": 2.0817,
|
| 542 |
+
"step": 44500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.48,
|
| 546 |
+
"learning_rate": 2.525814007191634e-05,
|
| 547 |
+
"loss": 2.0319,
|
| 548 |
+
"step": 45000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.5,
|
| 552 |
+
"learning_rate": 2.4983230517159856e-05,
|
| 553 |
+
"loss": 2.0231,
|
| 554 |
+
"step": 45500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 1.52,
|
| 558 |
+
"learning_rate": 2.4708320962403368e-05,
|
| 559 |
+
"loss": 2.0348,
|
| 560 |
+
"step": 46000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 1.53,
|
| 564 |
+
"learning_rate": 2.4433411407646887e-05,
|
| 565 |
+
"loss": 2.012,
|
| 566 |
+
"step": 46500
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 1.55,
|
| 570 |
+
"learning_rate": 2.41585018528904e-05,
|
| 571 |
+
"loss": 2.0662,
|
| 572 |
+
"step": 47000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.57,
|
| 576 |
+
"learning_rate": 2.3883592298133914e-05,
|
| 577 |
+
"loss": 2.0736,
|
| 578 |
+
"step": 47500
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 1.58,
|
| 582 |
+
"learning_rate": 2.360868274337743e-05,
|
| 583 |
+
"loss": 2.0574,
|
| 584 |
+
"step": 48000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.6,
|
| 588 |
+
"learning_rate": 2.3333773188620945e-05,
|
| 589 |
+
"loss": 2.1095,
|
| 590 |
+
"step": 48500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.62,
|
| 594 |
+
"learning_rate": 2.305886363386446e-05,
|
| 595 |
+
"loss": 2.0707,
|
| 596 |
+
"step": 49000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 1.63,
|
| 600 |
+
"learning_rate": 2.2783954079107976e-05,
|
| 601 |
+
"loss": 2.0368,
|
| 602 |
+
"step": 49500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.65,
|
| 606 |
+
"learning_rate": 2.250904452435149e-05,
|
| 607 |
+
"loss": 2.1298,
|
| 608 |
+
"step": 50000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.67,
|
| 612 |
+
"learning_rate": 2.2234134969595004e-05,
|
| 613 |
+
"loss": 2.0072,
|
| 614 |
+
"step": 50500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 1.68,
|
| 618 |
+
"learning_rate": 2.195922541483852e-05,
|
| 619 |
+
"loss": 1.9771,
|
| 620 |
+
"step": 51000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 1.7,
|
| 624 |
+
"learning_rate": 2.168431586008203e-05,
|
| 625 |
+
"loss": 1.9216,
|
| 626 |
+
"step": 51500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.72,
|
| 630 |
+
"learning_rate": 2.140940630532555e-05,
|
| 631 |
+
"loss": 1.9915,
|
| 632 |
+
"step": 52000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.73,
|
| 636 |
+
"learning_rate": 2.1134496750569062e-05,
|
| 637 |
+
"loss": 2.0122,
|
| 638 |
+
"step": 52500
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 1.75,
|
| 642 |
+
"learning_rate": 2.085958719581258e-05,
|
| 643 |
+
"loss": 2.0255,
|
| 644 |
+
"step": 53000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 1.76,
|
| 648 |
+
"learning_rate": 2.0584677641056093e-05,
|
| 649 |
+
"loss": 1.9464,
|
| 650 |
+
"step": 53500
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 1.78,
|
| 654 |
+
"learning_rate": 2.030976808629961e-05,
|
| 655 |
+
"loss": 1.9671,
|
| 656 |
+
"step": 54000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 1.8,
|
| 660 |
+
"learning_rate": 2.0034858531543124e-05,
|
| 661 |
+
"loss": 1.968,
|
| 662 |
+
"step": 54500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.81,
|
| 666 |
+
"learning_rate": 1.975994897678664e-05,
|
| 667 |
+
"loss": 2.0287,
|
| 668 |
+
"step": 55000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.83,
|
| 672 |
+
"learning_rate": 1.948503942203015e-05,
|
| 673 |
+
"loss": 1.9584,
|
| 674 |
+
"step": 55500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.85,
|
| 678 |
+
"learning_rate": 1.9210129867273667e-05,
|
| 679 |
+
"loss": 1.9247,
|
| 680 |
+
"step": 56000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 1.86,
|
| 684 |
+
"learning_rate": 1.8935220312517183e-05,
|
| 685 |
+
"loss": 1.9425,
|
| 686 |
+
"step": 56500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 1.88,
|
| 690 |
+
"learning_rate": 1.8660310757760698e-05,
|
| 691 |
+
"loss": 1.9221,
|
| 692 |
+
"step": 57000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 1.9,
|
| 696 |
+
"learning_rate": 1.8385401203004214e-05,
|
| 697 |
+
"loss": 1.9912,
|
| 698 |
+
"step": 57500
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 1.91,
|
| 702 |
+
"learning_rate": 1.8110491648247726e-05,
|
| 703 |
+
"loss": 1.9346,
|
| 704 |
+
"step": 58000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.93,
|
| 708 |
+
"learning_rate": 1.7835582093491245e-05,
|
| 709 |
+
"loss": 2.0059,
|
| 710 |
+
"step": 58500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.95,
|
| 714 |
+
"learning_rate": 1.7560672538734757e-05,
|
| 715 |
+
"loss": 1.9797,
|
| 716 |
+
"step": 59000
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.96,
|
| 720 |
+
"learning_rate": 1.7285762983978272e-05,
|
| 721 |
+
"loss": 2.023,
|
| 722 |
+
"step": 59500
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 1.98,
|
| 726 |
+
"learning_rate": 1.7010853429221788e-05,
|
| 727 |
+
"loss": 1.9738,
|
| 728 |
+
"step": 60000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 2.0,
|
| 732 |
+
"learning_rate": 1.6735943874465303e-05,
|
| 733 |
+
"loss": 1.951,
|
| 734 |
+
"step": 60500
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 2.01,
|
| 738 |
+
"learning_rate": 1.646103431970882e-05,
|
| 739 |
+
"loss": 1.8491,
|
| 740 |
+
"step": 61000
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 2.03,
|
| 744 |
+
"learning_rate": 1.618612476495233e-05,
|
| 745 |
+
"loss": 1.8754,
|
| 746 |
+
"step": 61500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 2.05,
|
| 750 |
+
"learning_rate": 1.5911215210195846e-05,
|
| 751 |
+
"loss": 1.856,
|
| 752 |
+
"step": 62000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 2.06,
|
| 756 |
+
"learning_rate": 1.563630565543936e-05,
|
| 757 |
+
"loss": 1.8521,
|
| 758 |
+
"step": 62500
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.08,
|
| 762 |
+
"learning_rate": 1.5361396100682877e-05,
|
| 763 |
+
"loss": 1.7808,
|
| 764 |
+
"step": 63000
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 2.09,
|
| 768 |
+
"learning_rate": 1.508648654592639e-05,
|
| 769 |
+
"loss": 1.8407,
|
| 770 |
+
"step": 63500
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 2.11,
|
| 774 |
+
"learning_rate": 1.4811576991169906e-05,
|
| 775 |
+
"loss": 1.85,
|
| 776 |
+
"step": 64000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 2.13,
|
| 780 |
+
"learning_rate": 1.453666743641342e-05,
|
| 781 |
+
"loss": 1.8037,
|
| 782 |
+
"step": 64500
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 2.14,
|
| 786 |
+
"learning_rate": 1.4261757881656937e-05,
|
| 787 |
+
"loss": 1.7651,
|
| 788 |
+
"step": 65000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 2.16,
|
| 792 |
+
"learning_rate": 1.3986848326900451e-05,
|
| 793 |
+
"loss": 1.8384,
|
| 794 |
+
"step": 65500
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 2.18,
|
| 798 |
+
"learning_rate": 1.3711938772143965e-05,
|
| 799 |
+
"loss": 1.8137,
|
| 800 |
+
"step": 66000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.19,
|
| 804 |
+
"learning_rate": 1.3437029217387482e-05,
|
| 805 |
+
"loss": 1.8169,
|
| 806 |
+
"step": 66500
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.21,
|
| 810 |
+
"learning_rate": 1.3162119662630996e-05,
|
| 811 |
+
"loss": 1.8162,
|
| 812 |
+
"step": 67000
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 2.23,
|
| 816 |
+
"learning_rate": 1.288721010787451e-05,
|
| 817 |
+
"loss": 1.8186,
|
| 818 |
+
"step": 67500
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 2.24,
|
| 822 |
+
"learning_rate": 1.2612300553118025e-05,
|
| 823 |
+
"loss": 1.707,
|
| 824 |
+
"step": 68000
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 2.26,
|
| 828 |
+
"learning_rate": 1.2337390998361539e-05,
|
| 829 |
+
"loss": 1.7471,
|
| 830 |
+
"step": 68500
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 2.28,
|
| 834 |
+
"learning_rate": 1.2062481443605054e-05,
|
| 835 |
+
"loss": 1.775,
|
| 836 |
+
"step": 69000
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 2.29,
|
| 840 |
+
"learning_rate": 1.178757188884857e-05,
|
| 841 |
+
"loss": 1.8094,
|
| 842 |
+
"step": 69500
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.31,
|
| 846 |
+
"learning_rate": 1.1512662334092083e-05,
|
| 847 |
+
"loss": 1.7498,
|
| 848 |
+
"step": 70000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 2.33,
|
| 852 |
+
"learning_rate": 1.1237752779335599e-05,
|
| 853 |
+
"loss": 1.8305,
|
| 854 |
+
"step": 70500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 2.34,
|
| 858 |
+
"learning_rate": 1.0962843224579114e-05,
|
| 859 |
+
"loss": 1.7538,
|
| 860 |
+
"step": 71000
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 2.36,
|
| 864 |
+
"learning_rate": 1.068793366982263e-05,
|
| 865 |
+
"loss": 1.7574,
|
| 866 |
+
"step": 71500
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 2.38,
|
| 870 |
+
"learning_rate": 1.0413024115066144e-05,
|
| 871 |
+
"loss": 1.7935,
|
| 872 |
+
"step": 72000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 2.39,
|
| 876 |
+
"learning_rate": 1.0138114560309659e-05,
|
| 877 |
+
"loss": 1.774,
|
| 878 |
+
"step": 72500
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.41,
|
| 882 |
+
"learning_rate": 9.863205005553173e-06,
|
| 883 |
+
"loss": 1.7352,
|
| 884 |
+
"step": 73000
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.42,
|
| 888 |
+
"learning_rate": 9.588295450796688e-06,
|
| 889 |
+
"loss": 1.7788,
|
| 890 |
+
"step": 73500
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 2.44,
|
| 894 |
+
"learning_rate": 9.313385896040202e-06,
|
| 895 |
+
"loss": 1.7613,
|
| 896 |
+
"step": 74000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 2.46,
|
| 900 |
+
"learning_rate": 9.038476341283717e-06,
|
| 901 |
+
"loss": 1.8059,
|
| 902 |
+
"step": 74500
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 2.47,
|
| 906 |
+
"learning_rate": 8.763566786527233e-06,
|
| 907 |
+
"loss": 1.7423,
|
| 908 |
+
"step": 75000
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 2.49,
|
| 912 |
+
"learning_rate": 8.488657231770748e-06,
|
| 913 |
+
"loss": 1.7693,
|
| 914 |
+
"step": 75500
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 2.51,
|
| 918 |
+
"learning_rate": 8.213747677014262e-06,
|
| 919 |
+
"loss": 1.776,
|
| 920 |
+
"step": 76000
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 2.52,
|
| 924 |
+
"learning_rate": 7.938838122257778e-06,
|
| 925 |
+
"loss": 1.7052,
|
| 926 |
+
"step": 76500
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.54,
|
| 930 |
+
"learning_rate": 7.663928567501293e-06,
|
| 931 |
+
"loss": 1.8079,
|
| 932 |
+
"step": 77000
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 2.56,
|
| 936 |
+
"learning_rate": 7.389019012744808e-06,
|
| 937 |
+
"loss": 1.7739,
|
| 938 |
+
"step": 77500
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 2.57,
|
| 942 |
+
"learning_rate": 7.1141094579883215e-06,
|
| 943 |
+
"loss": 1.6852,
|
| 944 |
+
"step": 78000
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 2.59,
|
| 948 |
+
"learning_rate": 6.839199903231837e-06,
|
| 949 |
+
"loss": 1.7038,
|
| 950 |
+
"step": 78500
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 2.61,
|
| 954 |
+
"learning_rate": 6.5642903484753525e-06,
|
| 955 |
+
"loss": 1.7107,
|
| 956 |
+
"step": 79000
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 2.62,
|
| 960 |
+
"learning_rate": 6.289380793718867e-06,
|
| 961 |
+
"loss": 1.7392,
|
| 962 |
+
"step": 79500
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 2.64,
|
| 966 |
+
"learning_rate": 6.014471238962382e-06,
|
| 967 |
+
"loss": 1.6904,
|
| 968 |
+
"step": 80000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.66,
|
| 972 |
+
"learning_rate": 5.739561684205896e-06,
|
| 973 |
+
"loss": 1.8101,
|
| 974 |
+
"step": 80500
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 2.67,
|
| 978 |
+
"learning_rate": 5.464652129449412e-06,
|
| 979 |
+
"loss": 1.6975,
|
| 980 |
+
"step": 81000
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 2.69,
|
| 984 |
+
"learning_rate": 5.1897425746929264e-06,
|
| 985 |
+
"loss": 1.7233,
|
| 986 |
+
"step": 81500
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 2.71,
|
| 990 |
+
"learning_rate": 4.914833019936441e-06,
|
| 991 |
+
"loss": 1.755,
|
| 992 |
+
"step": 82000
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 2.72,
|
| 996 |
+
"learning_rate": 4.639923465179956e-06,
|
| 997 |
+
"loss": 1.6656,
|
| 998 |
+
"step": 82500
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 2.74,
|
| 1002 |
+
"learning_rate": 4.365013910423471e-06,
|
| 1003 |
+
"loss": 1.7117,
|
| 1004 |
+
"step": 83000
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 2.75,
|
| 1008 |
+
"learning_rate": 4.090104355666986e-06,
|
| 1009 |
+
"loss": 1.7476,
|
| 1010 |
+
"step": 83500
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.77,
|
| 1014 |
+
"learning_rate": 3.815194800910501e-06,
|
| 1015 |
+
"loss": 1.7272,
|
| 1016 |
+
"step": 84000
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 2.79,
|
| 1020 |
+
"learning_rate": 3.5402852461540154e-06,
|
| 1021 |
+
"loss": 1.6628,
|
| 1022 |
+
"step": 84500
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 2.8,
|
| 1026 |
+
"learning_rate": 3.2653756913975305e-06,
|
| 1027 |
+
"loss": 1.7535,
|
| 1028 |
+
"step": 85000
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 2.82,
|
| 1032 |
+
"learning_rate": 2.990466136641045e-06,
|
| 1033 |
+
"loss": 1.68,
|
| 1034 |
+
"step": 85500
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 2.84,
|
| 1038 |
+
"learning_rate": 2.71555658188456e-06,
|
| 1039 |
+
"loss": 1.6641,
|
| 1040 |
+
"step": 86000
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 2.85,
|
| 1044 |
+
"learning_rate": 2.4406470271280748e-06,
|
| 1045 |
+
"loss": 1.6485,
|
| 1046 |
+
"step": 86500
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 2.87,
|
| 1050 |
+
"learning_rate": 2.16573747237159e-06,
|
| 1051 |
+
"loss": 1.6617,
|
| 1052 |
+
"step": 87000
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.89,
|
| 1056 |
+
"learning_rate": 1.8908279176151046e-06,
|
| 1057 |
+
"loss": 1.6926,
|
| 1058 |
+
"step": 87500
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 2.9,
|
| 1062 |
+
"learning_rate": 1.6159183628586195e-06,
|
| 1063 |
+
"loss": 1.6692,
|
| 1064 |
+
"step": 88000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 2.92,
|
| 1068 |
+
"learning_rate": 1.3410088081021345e-06,
|
| 1069 |
+
"loss": 1.734,
|
| 1070 |
+
"step": 88500
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 2.94,
|
| 1074 |
+
"learning_rate": 1.0660992533456494e-06,
|
| 1075 |
+
"loss": 1.6983,
|
| 1076 |
+
"step": 89000
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 2.95,
|
| 1080 |
+
"learning_rate": 7.911896985891643e-07,
|
| 1081 |
+
"loss": 1.6775,
|
| 1082 |
+
"step": 89500
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 2.97,
|
| 1086 |
+
"learning_rate": 5.16280143832679e-07,
|
| 1087 |
+
"loss": 1.6666,
|
| 1088 |
+
"step": 90000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 2.99,
|
| 1092 |
+
"learning_rate": 2.4137058907619396e-07,
|
| 1093 |
+
"loss": 1.7055,
|
| 1094 |
+
"step": 90500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 3.0,
|
| 1098 |
+
"step": 90939,
|
| 1099 |
+
"total_flos": 1.6358060886932256e+16,
|
| 1100 |
+
"train_loss": 1.3765672623247966,
|
| 1101 |
+
"train_runtime": 7341.0199,
|
| 1102 |
+
"train_samples_per_second": 99.099,
|
| 1103 |
+
"train_steps_per_second": 12.388
|
| 1104 |
+
}
|
| 1105 |
+
],
|
| 1106 |
+
"max_steps": 90939,
|
| 1107 |
+
"num_train_epochs": 3,
|
| 1108 |
+
"total_flos": 1.6358060886932256e+16,
|
| 1109 |
+
"trial_name": null,
|
| 1110 |
+
"trial_params": null
|
| 1111 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:18274da9c26e05fde73a998e5fa47cda2b6537f666a74e0db3f0219c9df7c0bb
|
| 3 |
+
size 2479
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|