Text Classification
Transformers
PyTorch
English
deberta-v2
Inference Endpoints
theblackcat102 commited on
Commit
c7ac1c6
1 Parent(s): e339002

Upload 8 files

Browse files
added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ {
2
+ "[MASK]": 128000
3
+ }
config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/deberta-v2-xxlarge",
3
+ "architectures": [
4
+ "DebertaV2ForSequenceClassification"
5
+ ],
6
+ "attention_head_size": 64,
7
+ "attention_probs_dropout_prob": 0.1,
8
+ "conv_act": "gelu",
9
+ "conv_kernel_size": 3,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 1536,
13
+ "id2label": {
14
+ "0": "LABEL_0"
15
+ },
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 6144,
18
+ "label2id": {
19
+ "LABEL_0": 0
20
+ },
21
+ "layer_norm_eps": 1e-07,
22
+ "max_position_embeddings": 512,
23
+ "max_relative_positions": -1,
24
+ "model_type": "deberta-v2",
25
+ "norm_rel_ebd": "layer_norm",
26
+ "num_attention_heads": 24,
27
+ "num_hidden_layers": 48,
28
+ "pad_token_id": 0,
29
+ "pooler_dropout": 0,
30
+ "pooler_hidden_act": "gelu",
31
+ "pooler_hidden_size": 1536,
32
+ "pos_att_type": [
33
+ "p2c",
34
+ "c2p"
35
+ ],
36
+ "position_biased_input": false,
37
+ "position_buckets": 256,
38
+ "problem_type": "regression",
39
+ "relative_attention": true,
40
+ "share_att_key": true,
41
+ "torch_dtype": "float16",
42
+ "transformers_version": "4.25.1",
43
+ "type_vocab_size": 0,
44
+ "vocab_size": 128100
45
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "[CLS]",
3
+ "cls_token": "[CLS]",
4
+ "eos_token": "[SEP]",
5
+ "mask_token": "[MASK]",
6
+ "pad_token": "[PAD]",
7
+ "sep_token": "[SEP]",
8
+ "unk_token": "[UNK]"
9
+ }
spm.model ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5598d5e96f339a8d980c15f9afd405a2e5e1be7db41de3ed13b0f03fac1e8c17
3
+ size 2447305
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "[CLS]",
3
+ "cls_token": "[CLS]",
4
+ "do_lower_case": false,
5
+ "eos_token": "[SEP]",
6
+ "mask_token": "[MASK]",
7
+ "model_max_length": 512,
8
+ "name_or_path": "microsoft/deberta-v2-xxlarge",
9
+ "pad_token": "[PAD]",
10
+ "sep_token": "[SEP]",
11
+ "sp_model_kwargs": {},
12
+ "special_tokens_map_file": null,
13
+ "split_by_punct": false,
14
+ "tokenizer_class": "DebertaV2Tokenizer",
15
+ "unk_token": "[UNK]",
16
+ "vocab_type": "spm"
17
+ }
trainer_state.json ADDED
@@ -0,0 +1,3616 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.923594785210515,
5
+ "global_step": 6000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.0,
12
+ "learning_rate": 2.6666666666666667e-08,
13
+ "loss": 0.6993,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 5.3333333333333334e-08,
19
+ "loss": 0.6971,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.01,
24
+ "learning_rate": 8.666666666666666e-08,
25
+ "loss": 0.7043,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.01,
30
+ "learning_rate": 1.2e-07,
31
+ "loss": 0.6975,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.02,
36
+ "learning_rate": 1.533333333333333e-07,
37
+ "loss": 0.699,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.02,
42
+ "learning_rate": 1.8666666666666667e-07,
43
+ "loss": 0.7028,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.02,
48
+ "learning_rate": 2.1666666666666667e-07,
49
+ "loss": 0.7,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.03,
54
+ "learning_rate": 2.5e-07,
55
+ "loss": 0.6964,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.03,
60
+ "learning_rate": 2.833333333333333e-07,
61
+ "loss": 0.7013,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.03,
66
+ "learning_rate": 3.166666666666666e-07,
67
+ "loss": 0.699,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.04,
72
+ "learning_rate": 3.4666666666666665e-07,
73
+ "loss": 0.7008,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.04,
78
+ "learning_rate": 3.7999999999999996e-07,
79
+ "loss": 0.6943,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.04,
84
+ "learning_rate": 4.1333333333333333e-07,
85
+ "loss": 0.7018,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.04,
90
+ "learning_rate": 4.4666666666666664e-07,
91
+ "loss": 0.6988,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.05,
96
+ "learning_rate": 4.8e-07,
97
+ "loss": 0.6998,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.05,
102
+ "learning_rate": 5.133333333333333e-07,
103
+ "loss": 0.696,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.05,
108
+ "learning_rate": 5.466666666666666e-07,
109
+ "loss": 0.6939,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.06,
114
+ "learning_rate": 5.766666666666666e-07,
115
+ "loss": 0.6959,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.06,
120
+ "learning_rate": 6.1e-07,
121
+ "loss": 0.7006,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.06,
126
+ "learning_rate": 6.433333333333332e-07,
127
+ "loss": 0.6958,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.07,
132
+ "learning_rate": 6.766666666666666e-07,
133
+ "loss": 0.6957,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.07,
138
+ "learning_rate": 7.1e-07,
139
+ "loss": 0.6925,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.07,
144
+ "learning_rate": 7.433333333333332e-07,
145
+ "loss": 0.6927,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.08,
150
+ "learning_rate": 7.766666666666666e-07,
151
+ "loss": 0.697,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.08,
156
+ "learning_rate": 8.1e-07,
157
+ "loss": 0.692,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.08,
162
+ "learning_rate": 8.433333333333333e-07,
163
+ "loss": 0.6998,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.09,
168
+ "learning_rate": 8.766666666666667e-07,
169
+ "loss": 0.6905,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.09,
174
+ "learning_rate": 9.1e-07,
175
+ "loss": 0.6925,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.09,
180
+ "learning_rate": 9.433333333333333e-07,
181
+ "loss": 0.699,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.1,
186
+ "learning_rate": 9.766666666666667e-07,
187
+ "loss": 0.6998,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.1,
192
+ "learning_rate": 1.0099999999999999e-06,
193
+ "loss": 0.6948,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.1,
198
+ "learning_rate": 1.0433333333333332e-06,
199
+ "loss": 0.7013,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.11,
204
+ "learning_rate": 1.0766666666666666e-06,
205
+ "loss": 0.6939,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.11,
210
+ "learning_rate": 1.11e-06,
211
+ "loss": 0.6937,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.11,
216
+ "learning_rate": 1.1433333333333333e-06,
217
+ "loss": 0.6921,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.12,
222
+ "learning_rate": 1.1766666666666667e-06,
223
+ "loss": 0.6917,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.12,
228
+ "learning_rate": 1.2099999999999998e-06,
229
+ "loss": 0.6898,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.12,
234
+ "learning_rate": 1.2433333333333334e-06,
235
+ "loss": 0.6929,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 0.13,
240
+ "learning_rate": 1.2766666666666666e-06,
241
+ "loss": 0.6988,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 0.13,
246
+ "learning_rate": 1.31e-06,
247
+ "loss": 0.6946,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 0.13,
252
+ "learning_rate": 1.3433333333333331e-06,
253
+ "loss": 0.6919,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 0.13,
258
+ "learning_rate": 1.3766666666666667e-06,
259
+ "loss": 0.6921,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 0.14,
264
+ "learning_rate": 1.4099999999999998e-06,
265
+ "loss": 0.6888,
266
+ "step": 430
267
+ },
268
+ {
269
+ "epoch": 0.14,
270
+ "learning_rate": 1.4433333333333332e-06,
271
+ "loss": 0.6912,
272
+ "step": 440
273
+ },
274
+ {
275
+ "epoch": 0.14,
276
+ "learning_rate": 1.4766666666666666e-06,
277
+ "loss": 0.6996,
278
+ "step": 450
279
+ },
280
+ {
281
+ "epoch": 0.15,
282
+ "learning_rate": 1.51e-06,
283
+ "loss": 0.6878,
284
+ "step": 460
285
+ },
286
+ {
287
+ "epoch": 0.15,
288
+ "learning_rate": 1.543333333333333e-06,
289
+ "loss": 0.6894,
290
+ "step": 470
291
+ },
292
+ {
293
+ "epoch": 0.15,
294
+ "learning_rate": 1.5766666666666667e-06,
295
+ "loss": 0.6808,
296
+ "step": 480
297
+ },
298
+ {
299
+ "epoch": 0.16,
300
+ "learning_rate": 1.6066666666666666e-06,
301
+ "loss": 0.6859,
302
+ "step": 490
303
+ },
304
+ {
305
+ "epoch": 0.16,
306
+ "learning_rate": 1.6399999999999998e-06,
307
+ "loss": 0.6766,
308
+ "step": 500
309
+ },
310
+ {
311
+ "epoch": 0.16,
312
+ "learning_rate": 1.6733333333333333e-06,
313
+ "loss": 0.6733,
314
+ "step": 510
315
+ },
316
+ {
317
+ "epoch": 0.17,
318
+ "learning_rate": 1.7066666666666667e-06,
319
+ "loss": 0.669,
320
+ "step": 520
321
+ },
322
+ {
323
+ "epoch": 0.17,
324
+ "learning_rate": 1.7399999999999999e-06,
325
+ "loss": 0.6586,
326
+ "step": 530
327
+ },
328
+ {
329
+ "epoch": 0.17,
330
+ "learning_rate": 1.7733333333333334e-06,
331
+ "loss": 0.6582,
332
+ "step": 540
333
+ },
334
+ {
335
+ "epoch": 0.18,
336
+ "learning_rate": 1.8066666666666666e-06,
337
+ "loss": 0.6755,
338
+ "step": 550
339
+ },
340
+ {
341
+ "epoch": 0.18,
342
+ "learning_rate": 1.84e-06,
343
+ "loss": 0.6702,
344
+ "step": 560
345
+ },
346
+ {
347
+ "epoch": 0.18,
348
+ "learning_rate": 1.8733333333333331e-06,
349
+ "loss": 0.6716,
350
+ "step": 570
351
+ },
352
+ {
353
+ "epoch": 0.19,
354
+ "learning_rate": 1.9066666666666667e-06,
355
+ "loss": 0.6565,
356
+ "step": 580
357
+ },
358
+ {
359
+ "epoch": 0.19,
360
+ "learning_rate": 1.94e-06,
361
+ "loss": 0.6626,
362
+ "step": 590
363
+ },
364
+ {
365
+ "epoch": 0.19,
366
+ "learning_rate": 1.9733333333333332e-06,
367
+ "loss": 0.6413,
368
+ "step": 600
369
+ },
370
+ {
371
+ "epoch": 0.2,
372
+ "learning_rate": 1.999999379017226e-06,
373
+ "loss": 0.6472,
374
+ "step": 610
375
+ },
376
+ {
377
+ "epoch": 0.2,
378
+ "learning_rate": 1.9999776447011173e-06,
379
+ "loss": 0.6428,
380
+ "step": 620
381
+ },
382
+ {
383
+ "epoch": 0.2,
384
+ "learning_rate": 1.9999248620175453e-06,
385
+ "loss": 0.66,
386
+ "step": 630
387
+ },
388
+ {
389
+ "epoch": 0.21,
390
+ "learning_rate": 1.9998508125827426e-06,
391
+ "loss": 0.6937,
392
+ "step": 640
393
+ },
394
+ {
395
+ "epoch": 0.21,
396
+ "learning_rate": 1.999739043312754e-06,
397
+ "loss": 0.6568,
398
+ "step": 650
399
+ },
400
+ {
401
+ "epoch": 0.21,
402
+ "learning_rate": 1.9995962330836514e-06,
403
+ "loss": 0.6446,
404
+ "step": 660
405
+ },
406
+ {
407
+ "epoch": 0.21,
408
+ "learning_rate": 1.9994223863295585e-06,
409
+ "loss": 0.6401,
410
+ "step": 670
411
+ },
412
+ {
413
+ "epoch": 0.22,
414
+ "learning_rate": 1.9992175084482535e-06,
415
+ "loss": 0.6356,
416
+ "step": 680
417
+ },
418
+ {
419
+ "epoch": 0.22,
420
+ "learning_rate": 1.998981605801002e-06,
421
+ "loss": 0.659,
422
+ "step": 690
423
+ },
424
+ {
425
+ "epoch": 0.22,
426
+ "learning_rate": 1.9987146857123613e-06,
427
+ "loss": 0.6346,
428
+ "step": 700
429
+ },
430
+ {
431
+ "epoch": 0.23,
432
+ "learning_rate": 1.9984167564699483e-06,
433
+ "loss": 0.6383,
434
+ "step": 710
435
+ },
436
+ {
437
+ "epoch": 0.23,
438
+ "learning_rate": 1.9980878273241866e-06,
439
+ "loss": 0.6394,
440
+ "step": 720
441
+ },
442
+ {
443
+ "epoch": 0.23,
444
+ "learning_rate": 1.997727908488018e-06,
445
+ "loss": 0.6344,
446
+ "step": 730
447
+ },
448
+ {
449
+ "epoch": 0.24,
450
+ "learning_rate": 1.997337011136584e-06,
451
+ "loss": 0.6226,
452
+ "step": 740
453
+ },
454
+ {
455
+ "epoch": 0.24,
456
+ "learning_rate": 1.9969151474068815e-06,
457
+ "loss": 0.6352,
458
+ "step": 750
459
+ },
460
+ {
461
+ "epoch": 0.24,
462
+ "learning_rate": 1.996462330397382e-06,
463
+ "loss": 0.6374,
464
+ "step": 760
465
+ },
466
+ {
467
+ "epoch": 0.25,
468
+ "learning_rate": 1.99597857416763e-06,
469
+ "loss": 0.6257,
470
+ "step": 770
471
+ },
472
+ {
473
+ "epoch": 0.25,
474
+ "learning_rate": 1.9954638937378017e-06,
475
+ "loss": 0.6307,
476
+ "step": 780
477
+ },
478
+ {
479
+ "epoch": 0.25,
480
+ "learning_rate": 1.9949183050882417e-06,
481
+ "loss": 0.6394,
482
+ "step": 790
483
+ },
484
+ {
485
+ "epoch": 0.26,
486
+ "learning_rate": 1.9943418251589657e-06,
487
+ "loss": 0.6472,
488
+ "step": 800
489
+ },
490
+ {
491
+ "epoch": 0.26,
492
+ "learning_rate": 1.9937344718491347e-06,
493
+ "loss": 0.6443,
494
+ "step": 810
495
+ },
496
+ {
497
+ "epoch": 0.26,
498
+ "learning_rate": 1.9930962640164985e-06,
499
+ "loss": 0.6225,
500
+ "step": 820
501
+ },
502
+ {
503
+ "epoch": 0.27,
504
+ "learning_rate": 1.9924272214768115e-06,
505
+ "loss": 0.6337,
506
+ "step": 830
507
+ },
508
+ {
509
+ "epoch": 0.27,
510
+ "learning_rate": 1.991727365003217e-06,
511
+ "loss": 0.6136,
512
+ "step": 840
513
+ },
514
+ {
515
+ "epoch": 0.27,
516
+ "learning_rate": 1.9909967163256016e-06,
517
+ "loss": 0.6309,
518
+ "step": 850
519
+ },
520
+ {
521
+ "epoch": 0.28,
522
+ "learning_rate": 1.9902352981299204e-06,
523
+ "loss": 0.6311,
524
+ "step": 860
525
+ },
526
+ {
527
+ "epoch": 0.28,
528
+ "learning_rate": 1.9894431340574953e-06,
529
+ "loss": 0.6111,
530
+ "step": 870
531
+ },
532
+ {
533
+ "epoch": 0.28,
534
+ "learning_rate": 1.9886202487042762e-06,
535
+ "loss": 0.6334,
536
+ "step": 880
537
+ },
538
+ {
539
+ "epoch": 0.29,
540
+ "learning_rate": 1.9877666676200812e-06,
541
+ "loss": 0.6193,
542
+ "step": 890
543
+ },
544
+ {
545
+ "epoch": 0.29,
546
+ "learning_rate": 1.9868824173078025e-06,
547
+ "loss": 0.6369,
548
+ "step": 900
549
+ },
550
+ {
551
+ "epoch": 0.29,
552
+ "learning_rate": 1.9859675252225824e-06,
553
+ "loss": 0.6283,
554
+ "step": 910
555
+ },
556
+ {
557
+ "epoch": 0.29,
558
+ "learning_rate": 1.985022019770961e-06,
559
+ "loss": 0.5903,
560
+ "step": 920
561
+ },
562
+ {
563
+ "epoch": 0.3,
564
+ "learning_rate": 1.9840459303099958e-06,
565
+ "loss": 0.6118,
566
+ "step": 930
567
+ },
568
+ {
569
+ "epoch": 0.3,
570
+ "learning_rate": 1.983039287146348e-06,
571
+ "loss": 0.6102,
572
+ "step": 940
573
+ },
574
+ {
575
+ "epoch": 0.3,
576
+ "learning_rate": 1.982002121535344e-06,
577
+ "loss": 0.6058,
578
+ "step": 950
579
+ },
580
+ {
581
+ "epoch": 0.31,
582
+ "learning_rate": 1.9809344656800026e-06,
583
+ "loss": 0.6103,
584
+ "step": 960
585
+ },
586
+ {
587
+ "epoch": 0.31,
588
+ "learning_rate": 1.9798363527300355e-06,
589
+ "loss": 0.6137,
590
+ "step": 970
591
+ },
592
+ {
593
+ "epoch": 0.31,
594
+ "learning_rate": 1.97870781678082e-06,
595
+ "loss": 0.6079,
596
+ "step": 980
597
+ },
598
+ {
599
+ "epoch": 0.32,
600
+ "learning_rate": 1.9775488928723377e-06,
601
+ "loss": 0.5953,
602
+ "step": 990
603
+ },
604
+ {
605
+ "epoch": 0.32,
606
+ "learning_rate": 1.976359616988089e-06,
607
+ "loss": 0.5808,
608
+ "step": 1000
609
+ },
610
+ {
611
+ "epoch": 0.32,
612
+ "learning_rate": 1.975140026053974e-06,
613
+ "loss": 0.604,
614
+ "step": 1010
615
+ },
616
+ {
617
+ "epoch": 0.33,
618
+ "learning_rate": 1.9738901579371454e-06,
619
+ "loss": 0.6013,
620
+ "step": 1020
621
+ },
622
+ {
623
+ "epoch": 0.33,
624
+ "learning_rate": 1.9726100514448373e-06,
625
+ "loss": 0.5625,
626
+ "step": 1030
627
+ },
628
+ {
629
+ "epoch": 0.33,
630
+ "learning_rate": 1.9712997463231535e-06,
631
+ "loss": 0.5886,
632
+ "step": 1040
633
+ },
634
+ {
635
+ "epoch": 0.34,
636
+ "learning_rate": 1.9699592832558395e-06,
637
+ "loss": 0.5952,
638
+ "step": 1050
639
+ },
640
+ {
641
+ "epoch": 0.34,
642
+ "learning_rate": 1.968588703863015e-06,
643
+ "loss": 0.5767,
644
+ "step": 1060
645
+ },
646
+ {
647
+ "epoch": 0.34,
648
+ "learning_rate": 1.967188050699884e-06,
649
+ "loss": 0.5858,
650
+ "step": 1070
651
+ },
652
+ {
653
+ "epoch": 0.35,
654
+ "learning_rate": 1.9657573672554125e-06,
655
+ "loss": 0.5786,
656
+ "step": 1080
657
+ },
658
+ {
659
+ "epoch": 0.35,
660
+ "learning_rate": 1.9642966979509797e-06,
661
+ "loss": 0.5701,
662
+ "step": 1090
663
+ },
664
+ {
665
+ "epoch": 0.35,
666
+ "learning_rate": 1.962806088138996e-06,
667
+ "loss": 0.5874,
668
+ "step": 1100
669
+ },
670
+ {
671
+ "epoch": 0.36,
672
+ "learning_rate": 1.9612855841014974e-06,
673
+ "loss": 0.603,
674
+ "step": 1110
675
+ },
676
+ {
677
+ "epoch": 0.36,
678
+ "learning_rate": 1.959735233048708e-06,
679
+ "loss": 0.5846,
680
+ "step": 1120
681
+ },
682
+ {
683
+ "epoch": 0.36,
684
+ "learning_rate": 1.9581550831175724e-06,
685
+ "loss": 0.5769,
686
+ "step": 1130
687
+ },
688
+ {
689
+ "epoch": 0.37,
690
+ "learning_rate": 1.956545183370263e-06,
691
+ "loss": 0.5643,
692
+ "step": 1140
693
+ },
694
+ {
695
+ "epoch": 0.37,
696
+ "learning_rate": 1.954905583792657e-06,
697
+ "loss": 0.59,
698
+ "step": 1150
699
+ },
700
+ {
701
+ "epoch": 0.37,
702
+ "learning_rate": 1.953236335292782e-06,
703
+ "loss": 0.5629,
704
+ "step": 1160
705
+ },
706
+ {
707
+ "epoch": 0.38,
708
+ "learning_rate": 1.9515374896992374e-06,
709
+ "loss": 0.5685,
710
+ "step": 1170
711
+ },
712
+ {
713
+ "epoch": 0.38,
714
+ "learning_rate": 1.949809099759585e-06,
715
+ "loss": 0.5782,
716
+ "step": 1180
717
+ },
718
+ {
719
+ "epoch": 0.38,
720
+ "learning_rate": 1.9480512191387103e-06,
721
+ "loss": 0.578,
722
+ "step": 1190
723
+ },
724
+ {
725
+ "epoch": 0.38,
726
+ "learning_rate": 1.946263902417157e-06,
727
+ "loss": 0.5844,
728
+ "step": 1200
729
+ },
730
+ {
731
+ "epoch": 0.39,
732
+ "learning_rate": 1.944447205089432e-06,
733
+ "loss": 0.5617,
734
+ "step": 1210
735
+ },
736
+ {
737
+ "epoch": 0.39,
738
+ "learning_rate": 1.942601183562282e-06,
739
+ "loss": 0.5909,
740
+ "step": 1220
741
+ },
742
+ {
743
+ "epoch": 0.39,
744
+ "learning_rate": 1.940725895152944e-06,
745
+ "loss": 0.5597,
746
+ "step": 1230
747
+ },
748
+ {
749
+ "epoch": 0.4,
750
+ "learning_rate": 1.938821398087363e-06,
751
+ "loss": 0.5374,
752
+ "step": 1240
753
+ },
754
+ {
755
+ "epoch": 0.4,
756
+ "learning_rate": 1.9368877514983846e-06,
757
+ "loss": 0.5788,
758
+ "step": 1250
759
+ },
760
+ {
761
+ "epoch": 0.4,
762
+ "learning_rate": 1.9349250154239227e-06,
763
+ "loss": 0.5733,
764
+ "step": 1260
765
+ },
766
+ {
767
+ "epoch": 0.41,
768
+ "learning_rate": 1.932933250805089e-06,
769
+ "loss": 0.5579,
770
+ "step": 1270
771
+ },
772
+ {
773
+ "epoch": 0.41,
774
+ "learning_rate": 1.9309125194843065e-06,
775
+ "loss": 0.5536,
776
+ "step": 1280
777
+ },
778
+ {
779
+ "epoch": 0.41,
780
+ "learning_rate": 1.9290691466029904e-06,
781
+ "loss": 0.5624,
782
+ "step": 1290
783
+ },
784
+ {
785
+ "epoch": 0.42,
786
+ "learning_rate": 1.9269935521439704e-06,
787
+ "loss": 0.5594,
788
+ "step": 1300
789
+ },
790
+ {
791
+ "epoch": 0.42,
792
+ "learning_rate": 1.924889175405071e-06,
793
+ "loss": 0.6108,
794
+ "step": 1310
795
+ },
796
+ {
797
+ "epoch": 0.42,
798
+ "learning_rate": 1.9227560817252142e-06,
799
+ "loss": 0.5796,
800
+ "step": 1320
801
+ },
802
+ {
803
+ "epoch": 0.43,
804
+ "learning_rate": 1.9205943373349583e-06,
805
+ "loss": 0.5579,
806
+ "step": 1330
807
+ },
808
+ {
809
+ "epoch": 0.43,
810
+ "learning_rate": 1.9184040093544373e-06,
811
+ "loss": 0.5476,
812
+ "step": 1340
813
+ },
814
+ {
815
+ "epoch": 0.43,
816
+ "learning_rate": 1.9161851657912802e-06,
817
+ "loss": 0.551,
818
+ "step": 1350
819
+ },
820
+ {
821
+ "epoch": 0.44,
822
+ "learning_rate": 1.9139378755384964e-06,
823
+ "loss": 0.5362,
824
+ "step": 1360
825
+ },
826
+ {
827
+ "epoch": 0.44,
828
+ "learning_rate": 1.9116622083723404e-06,
829
+ "loss": 0.5626,
830
+ "step": 1370
831
+ },
832
+ {
833
+ "epoch": 0.44,
834
+ "learning_rate": 1.909358234950141e-06,
835
+ "loss": 0.5954,
836
+ "step": 1380
837
+ },
838
+ {
839
+ "epoch": 0.45,
840
+ "learning_rate": 1.9070260268081125e-06,
841
+ "loss": 0.5523,
842
+ "step": 1390
843
+ },
844
+ {
845
+ "epoch": 0.45,
846
+ "learning_rate": 1.9046656563591275e-06,
847
+ "loss": 0.5732,
848
+ "step": 1400
849
+ },
850
+ {
851
+ "epoch": 0.45,
852
+ "learning_rate": 1.9022771968904741e-06,
853
+ "loss": 0.5865,
854
+ "step": 1410
855
+ },
856
+ {
857
+ "epoch": 0.46,
858
+ "learning_rate": 1.8998607225615772e-06,
859
+ "loss": 0.5725,
860
+ "step": 1420
861
+ },
862
+ {
863
+ "epoch": 0.46,
864
+ "learning_rate": 1.8974163084016972e-06,
865
+ "loss": 0.5925,
866
+ "step": 1430
867
+ },
868
+ {
869
+ "epoch": 0.46,
870
+ "learning_rate": 1.8949440303076001e-06,
871
+ "loss": 0.5535,
872
+ "step": 1440
873
+ },
874
+ {
875
+ "epoch": 0.46,
876
+ "learning_rate": 1.8924439650412001e-06,
877
+ "loss": 0.5363,
878
+ "step": 1450
879
+ },
880
+ {
881
+ "epoch": 0.47,
882
+ "learning_rate": 1.8899161902271778e-06,
883
+ "loss": 0.5486,
884
+ "step": 1460
885
+ },
886
+ {
887
+ "epoch": 0.47,
888
+ "learning_rate": 1.8873607843505695e-06,
889
+ "loss": 0.5412,
890
+ "step": 1470
891
+ },
892
+ {
893
+ "epoch": 0.47,
894
+ "learning_rate": 1.884777826754329e-06,
895
+ "loss": 0.577,
896
+ "step": 1480
897
+ },
898
+ {
899
+ "epoch": 0.48,
900
+ "learning_rate": 1.8821673976368662e-06,
901
+ "loss": 0.5408,
902
+ "step": 1490
903
+ },
904
+ {
905
+ "epoch": 0.48,
906
+ "learning_rate": 1.8795295780495551e-06,
907
+ "loss": 0.5521,
908
+ "step": 1500
909
+ },
910
+ {
911
+ "epoch": 0.48,
912
+ "learning_rate": 1.876864449894219e-06,
913
+ "loss": 0.5774,
914
+ "step": 1510
915
+ },
916
+ {
917
+ "epoch": 0.49,
918
+ "learning_rate": 1.8741720959205854e-06,
919
+ "loss": 0.5678,
920
+ "step": 1520
921
+ },
922
+ {
923
+ "epoch": 0.49,
924
+ "learning_rate": 1.8714525997237192e-06,
925
+ "loss": 0.5533,
926
+ "step": 1530
927
+ },
928
+ {
929
+ "epoch": 0.49,
930
+ "learning_rate": 1.8687060457414253e-06,
931
+ "loss": 0.5541,
932
+ "step": 1540
933
+ },
934
+ {
935
+ "epoch": 0.5,
936
+ "learning_rate": 1.8659325192516267e-06,
937
+ "loss": 0.5594,
938
+ "step": 1550
939
+ },
940
+ {
941
+ "epoch": 0.5,
942
+ "learning_rate": 1.8631321063697189e-06,
943
+ "loss": 0.5485,
944
+ "step": 1560
945
+ },
946
+ {
947
+ "epoch": 0.5,
948
+ "learning_rate": 1.8603048940458935e-06,
949
+ "loss": 0.5777,
950
+ "step": 1570
951
+ },
952
+ {
953
+ "epoch": 0.51,
954
+ "learning_rate": 1.8574509700624403e-06,
955
+ "loss": 0.5569,
956
+ "step": 1580
957
+ },
958
+ {
959
+ "epoch": 0.51,
960
+ "learning_rate": 1.8545704230310206e-06,
961
+ "loss": 0.5611,
962
+ "step": 1590
963
+ },
964
+ {
965
+ "epoch": 0.51,
966
+ "learning_rate": 1.8516633423899166e-06,
967
+ "loss": 0.5451,
968
+ "step": 1600
969
+ },
970
+ {
971
+ "epoch": 0.52,
972
+ "learning_rate": 1.8487298184012543e-06,
973
+ "loss": 0.5906,
974
+ "step": 1610
975
+ },
976
+ {
977
+ "epoch": 0.52,
978
+ "learning_rate": 1.8457699421482003e-06,
979
+ "loss": 0.555,
980
+ "step": 1620
981
+ },
982
+ {
983
+ "epoch": 0.52,
984
+ "learning_rate": 1.8427838055321349e-06,
985
+ "loss": 0.5435,
986
+ "step": 1630
987
+ },
988
+ {
989
+ "epoch": 0.53,
990
+ "learning_rate": 1.8397715012697976e-06,
991
+ "loss": 0.5789,
992
+ "step": 1640
993
+ },
994
+ {
995
+ "epoch": 0.53,
996
+ "learning_rate": 1.836733122890409e-06,
997
+ "loss": 0.56,
998
+ "step": 1650
999
+ },
1000
+ {
1001
+ "epoch": 0.53,
1002
+ "learning_rate": 1.8336687647327666e-06,
1003
+ "loss": 0.545,
1004
+ "step": 1660
1005
+ },
1006
+ {
1007
+ "epoch": 0.54,
1008
+ "learning_rate": 1.8305785219423155e-06,
1009
+ "loss": 0.5888,
1010
+ "step": 1670
1011
+ },
1012
+ {
1013
+ "epoch": 0.54,
1014
+ "learning_rate": 1.827462490468194e-06,
1015
+ "loss": 0.5551,
1016
+ "step": 1680
1017
+ },
1018
+ {
1019
+ "epoch": 0.54,
1020
+ "learning_rate": 1.824320767060256e-06,
1021
+ "loss": 0.5491,
1022
+ "step": 1690
1023
+ },
1024
+ {
1025
+ "epoch": 0.54,
1026
+ "learning_rate": 1.821153449266064e-06,
1027
+ "loss": 0.5504,
1028
+ "step": 1700
1029
+ },
1030
+ {
1031
+ "epoch": 0.55,
1032
+ "learning_rate": 1.817960635427864e-06,
1033
+ "loss": 0.5411,
1034
+ "step": 1710
1035
+ },
1036
+ {
1037
+ "epoch": 0.55,
1038
+ "learning_rate": 1.8150653857736957e-06,
1039
+ "loss": 0.5527,
1040
+ "step": 1720
1041
+ },
1042
+ {
1043
+ "epoch": 0.55,
1044
+ "learning_rate": 1.8118244032175022e-06,
1045
+ "loss": 0.5753,
1046
+ "step": 1730
1047
+ },
1048
+ {
1049
+ "epoch": 0.56,
1050
+ "learning_rate": 1.8085582142754256e-06,
1051
+ "loss": 0.5519,
1052
+ "step": 1740
1053
+ },
1054
+ {
1055
+ "epoch": 0.56,
1056
+ "learning_rate": 1.8052669203595676e-06,
1057
+ "loss": 0.5601,
1058
+ "step": 1750
1059
+ },
1060
+ {
1061
+ "epoch": 0.56,
1062
+ "learning_rate": 1.8019506236615156e-06,
1063
+ "loss": 0.5657,
1064
+ "step": 1760
1065
+ },
1066
+ {
1067
+ "epoch": 0.57,
1068
+ "learning_rate": 1.79860942714917e-06,
1069
+ "loss": 0.5304,
1070
+ "step": 1770
1071
+ },
1072
+ {
1073
+ "epoch": 0.57,
1074
+ "learning_rate": 1.7952434345635467e-06,
1075
+ "loss": 0.5566,
1076
+ "step": 1780
1077
+ },
1078
+ {
1079
+ "epoch": 0.57,
1080
+ "learning_rate": 1.7918527504155572e-06,
1081
+ "loss": 0.5712,
1082
+ "step": 1790
1083
+ },
1084
+ {
1085
+ "epoch": 0.58,
1086
+ "learning_rate": 1.7884374799827624e-06,
1087
+ "loss": 0.541,
1088
+ "step": 1800
1089
+ },
1090
+ {
1091
+ "epoch": 0.58,
1092
+ "learning_rate": 1.7849977293061041e-06,
1093
+ "loss": 0.5325,
1094
+ "step": 1810
1095
+ },
1096
+ {
1097
+ "epoch": 0.58,
1098
+ "learning_rate": 1.7815336051866126e-06,
1099
+ "loss": 0.5566,
1100
+ "step": 1820
1101
+ },
1102
+ {
1103
+ "epoch": 0.59,
1104
+ "learning_rate": 1.7780452151820913e-06,
1105
+ "loss": 0.5507,
1106
+ "step": 1830
1107
+ },
1108
+ {
1109
+ "epoch": 0.59,
1110
+ "learning_rate": 1.7745326676037764e-06,
1111
+ "loss": 0.5756,
1112
+ "step": 1840
1113
+ },
1114
+ {
1115
+ "epoch": 0.59,
1116
+ "learning_rate": 1.7709960715129736e-06,
1117
+ "loss": 0.545,
1118
+ "step": 1850
1119
+ },
1120
+ {
1121
+ "epoch": 0.6,
1122
+ "learning_rate": 1.7674355367176728e-06,
1123
+ "loss": 0.535,
1124
+ "step": 1860
1125
+ },
1126
+ {
1127
+ "epoch": 0.6,
1128
+ "learning_rate": 1.7638511737691383e-06,
1129
+ "loss": 0.5461,
1130
+ "step": 1870
1131
+ },
1132
+ {
1133
+ "epoch": 0.6,
1134
+ "learning_rate": 1.7602430939584756e-06,
1135
+ "loss": 0.5583,
1136
+ "step": 1880
1137
+ },
1138
+ {
1139
+ "epoch": 0.61,
1140
+ "learning_rate": 1.7566114093131773e-06,
1141
+ "loss": 0.5313,
1142
+ "step": 1890
1143
+ },
1144
+ {
1145
+ "epoch": 0.61,
1146
+ "learning_rate": 1.7529562325936433e-06,
1147
+ "loss": 0.5618,
1148
+ "step": 1900
1149
+ },
1150
+ {
1151
+ "epoch": 0.61,
1152
+ "learning_rate": 1.749277677289681e-06,
1153
+ "loss": 0.5486,
1154
+ "step": 1910
1155
+ },
1156
+ {
1157
+ "epoch": 0.62,
1158
+ "learning_rate": 1.7455758576169797e-06,
1159
+ "loss": 0.5512,
1160
+ "step": 1920
1161
+ },
1162
+ {
1163
+ "epoch": 0.62,
1164
+ "learning_rate": 1.7418508885135672e-06,
1165
+ "loss": 0.531,
1166
+ "step": 1930
1167
+ },
1168
+ {
1169
+ "epoch": 0.62,
1170
+ "learning_rate": 1.7381028856362383e-06,
1171
+ "loss": 0.546,
1172
+ "step": 1940
1173
+ },
1174
+ {
1175
+ "epoch": 0.63,
1176
+ "learning_rate": 1.734331965356965e-06,
1177
+ "loss": 0.5604,
1178
+ "step": 1950
1179
+ },
1180
+ {
1181
+ "epoch": 0.63,
1182
+ "learning_rate": 1.7305382447592835e-06,
1183
+ "loss": 0.5492,
1184
+ "step": 1960
1185
+ },
1186
+ {
1187
+ "epoch": 0.63,
1188
+ "learning_rate": 1.7267218416346581e-06,
1189
+ "loss": 0.5331,
1190
+ "step": 1970
1191
+ },
1192
+ {
1193
+ "epoch": 0.63,
1194
+ "learning_rate": 1.722882874478824e-06,
1195
+ "loss": 0.5516,
1196
+ "step": 1980
1197
+ },
1198
+ {
1199
+ "epoch": 0.64,
1200
+ "learning_rate": 1.7190214624881096e-06,
1201
+ "loss": 0.551,
1202
+ "step": 1990
1203
+ },
1204
+ {
1205
+ "epoch": 0.64,
1206
+ "learning_rate": 1.7151377255557329e-06,
1207
+ "loss": 0.5769,
1208
+ "step": 2000
1209
+ },
1210
+ {
1211
+ "epoch": 0.64,
1212
+ "learning_rate": 1.7112317842680812e-06,
1213
+ "loss": 0.555,
1214
+ "step": 2010
1215
+ },
1216
+ {
1217
+ "epoch": 0.65,
1218
+ "learning_rate": 1.7073037599009666e-06,
1219
+ "loss": 0.5256,
1220
+ "step": 2020
1221
+ },
1222
+ {
1223
+ "epoch": 0.65,
1224
+ "learning_rate": 1.7033537744158588e-06,
1225
+ "loss": 0.5507,
1226
+ "step": 2030
1227
+ },
1228
+ {
1229
+ "epoch": 0.65,
1230
+ "learning_rate": 1.6993819504561003e-06,
1231
+ "loss": 0.5588,
1232
+ "step": 2040
1233
+ },
1234
+ {
1235
+ "epoch": 0.66,
1236
+ "learning_rate": 1.6953884113430984e-06,
1237
+ "loss": 0.5347,
1238
+ "step": 2050
1239
+ },
1240
+ {
1241
+ "epoch": 0.66,
1242
+ "learning_rate": 1.6913732810724946e-06,
1243
+ "loss": 0.5565,
1244
+ "step": 2060
1245
+ },
1246
+ {
1247
+ "epoch": 0.66,
1248
+ "learning_rate": 1.687336684310316e-06,
1249
+ "loss": 0.5568,
1250
+ "step": 2070
1251
+ },
1252
+ {
1253
+ "epoch": 0.67,
1254
+ "learning_rate": 1.683278746389104e-06,
1255
+ "loss": 0.543,
1256
+ "step": 2080
1257
+ },
1258
+ {
1259
+ "epoch": 0.67,
1260
+ "learning_rate": 1.6791995933040228e-06,
1261
+ "loss": 0.5402,
1262
+ "step": 2090
1263
+ },
1264
+ {
1265
+ "epoch": 0.67,
1266
+ "learning_rate": 1.675099351708948e-06,
1267
+ "loss": 0.5536,
1268
+ "step": 2100
1269
+ },
1270
+ {
1271
+ "epoch": 0.68,
1272
+ "learning_rate": 1.670978148912533e-06,
1273
+ "loss": 0.5522,
1274
+ "step": 2110
1275
+ },
1276
+ {
1277
+ "epoch": 0.68,
1278
+ "learning_rate": 1.6668361128742574e-06,
1279
+ "loss": 0.5611,
1280
+ "step": 2120
1281
+ },
1282
+ {
1283
+ "epoch": 0.68,
1284
+ "learning_rate": 1.6626733722004533e-06,
1285
+ "loss": 0.5365,
1286
+ "step": 2130
1287
+ },
1288
+ {
1289
+ "epoch": 0.69,
1290
+ "learning_rate": 1.6584900561403119e-06,
1291
+ "loss": 0.5366,
1292
+ "step": 2140
1293
+ },
1294
+ {
1295
+ "epoch": 0.69,
1296
+ "learning_rate": 1.6542862945818716e-06,
1297
+ "loss": 0.5411,
1298
+ "step": 2150
1299
+ },
1300
+ {
1301
+ "epoch": 0.69,
1302
+ "learning_rate": 1.6500622180479832e-06,
1303
+ "loss": 0.5077,
1304
+ "step": 2160
1305
+ },
1306
+ {
1307
+ "epoch": 0.7,
1308
+ "learning_rate": 1.64581795769226e-06,
1309
+ "loss": 0.5425,
1310
+ "step": 2170
1311
+ },
1312
+ {
1313
+ "epoch": 0.7,
1314
+ "learning_rate": 1.6415536452950023e-06,
1315
+ "loss": 0.5599,
1316
+ "step": 2180
1317
+ },
1318
+ {
1319
+ "epoch": 0.7,
1320
+ "learning_rate": 1.6372694132591085e-06,
1321
+ "loss": 0.5502,
1322
+ "step": 2190
1323
+ },
1324
+ {
1325
+ "epoch": 0.71,
1326
+ "learning_rate": 1.6329653946059633e-06,
1327
+ "loss": 0.5308,
1328
+ "step": 2200
1329
+ },
1330
+ {
1331
+ "epoch": 0.71,
1332
+ "learning_rate": 1.6286417229713061e-06,
1333
+ "loss": 0.522,
1334
+ "step": 2210
1335
+ },
1336
+ {
1337
+ "epoch": 0.71,
1338
+ "learning_rate": 1.6242985326010845e-06,
1339
+ "loss": 0.5391,
1340
+ "step": 2220
1341
+ },
1342
+ {
1343
+ "epoch": 0.71,
1344
+ "learning_rate": 1.6199359583472832e-06,
1345
+ "loss": 0.5216,
1346
+ "step": 2230
1347
+ },
1348
+ {
1349
+ "epoch": 0.72,
1350
+ "learning_rate": 1.6155541356637393e-06,
1351
+ "loss": 0.5354,
1352
+ "step": 2240
1353
+ },
1354
+ {
1355
+ "epoch": 0.72,
1356
+ "learning_rate": 1.6111532006019346e-06,
1357
+ "loss": 0.5279,
1358
+ "step": 2250
1359
+ },
1360
+ {
1361
+ "epoch": 0.72,
1362
+ "learning_rate": 1.6067332898067732e-06,
1363
+ "loss": 0.5437,
1364
+ "step": 2260
1365
+ },
1366
+ {
1367
+ "epoch": 0.73,
1368
+ "learning_rate": 1.6022945405123372e-06,
1369
+ "loss": 0.5246,
1370
+ "step": 2270
1371
+ },
1372
+ {
1373
+ "epoch": 0.73,
1374
+ "learning_rate": 1.5978370905376271e-06,
1375
+ "loss": 0.5445,
1376
+ "step": 2280
1377
+ },
1378
+ {
1379
+ "epoch": 0.73,
1380
+ "learning_rate": 1.5933610782822815e-06,
1381
+ "loss": 0.5355,
1382
+ "step": 2290
1383
+ },
1384
+ {
1385
+ "epoch": 0.74,
1386
+ "learning_rate": 1.5888666427222805e-06,
1387
+ "loss": 0.5239,
1388
+ "step": 2300
1389
+ },
1390
+ {
1391
+ "epoch": 0.74,
1392
+ "learning_rate": 1.5843539234056306e-06,
1393
+ "loss": 0.5275,
1394
+ "step": 2310
1395
+ },
1396
+ {
1397
+ "epoch": 0.74,
1398
+ "learning_rate": 1.5798230604480316e-06,
1399
+ "loss": 0.5415,
1400
+ "step": 2320
1401
+ },
1402
+ {
1403
+ "epoch": 0.75,
1404
+ "learning_rate": 1.5752741945285263e-06,
1405
+ "loss": 0.5405,
1406
+ "step": 2330
1407
+ },
1408
+ {
1409
+ "epoch": 0.75,
1410
+ "learning_rate": 1.5707074668851332e-06,
1411
+ "loss": 0.5397,
1412
+ "step": 2340
1413
+ },
1414
+ {
1415
+ "epoch": 0.75,
1416
+ "learning_rate": 1.56612301931046e-06,
1417
+ "loss": 0.5301,
1418
+ "step": 2350
1419
+ },
1420
+ {
1421
+ "epoch": 0.76,
1422
+ "learning_rate": 1.5615209941473013e-06,
1423
+ "loss": 0.5678,
1424
+ "step": 2360
1425
+ },
1426
+ {
1427
+ "epoch": 0.76,
1428
+ "learning_rate": 1.5569015342842201e-06,
1429
+ "loss": 0.5284,
1430
+ "step": 2370
1431
+ },
1432
+ {
1433
+ "epoch": 0.76,
1434
+ "learning_rate": 1.5522647831511102e-06,
1435
+ "loss": 0.529,
1436
+ "step": 2380
1437
+ },
1438
+ {
1439
+ "epoch": 0.77,
1440
+ "learning_rate": 1.5476108847147428e-06,
1441
+ "loss": 0.5362,
1442
+ "step": 2390
1443
+ },
1444
+ {
1445
+ "epoch": 0.77,
1446
+ "learning_rate": 1.5429399834742971e-06,
1447
+ "loss": 0.5337,
1448
+ "step": 2400
1449
+ },
1450
+ {
1451
+ "epoch": 0.77,
1452
+ "learning_rate": 1.5382522244568733e-06,
1453
+ "loss": 0.5356,
1454
+ "step": 2410
1455
+ },
1456
+ {
1457
+ "epoch": 0.78,
1458
+ "learning_rate": 1.5335477532129902e-06,
1459
+ "loss": 0.5174,
1460
+ "step": 2420
1461
+ },
1462
+ {
1463
+ "epoch": 0.78,
1464
+ "learning_rate": 1.5288267158120642e-06,
1465
+ "loss": 0.5387,
1466
+ "step": 2430
1467
+ },
1468
+ {
1469
+ "epoch": 0.78,
1470
+ "learning_rate": 1.5240892588378773e-06,
1471
+ "loss": 0.5443,
1472
+ "step": 2440
1473
+ },
1474
+ {
1475
+ "epoch": 0.79,
1476
+ "learning_rate": 1.5193355293840217e-06,
1477
+ "loss": 0.534,
1478
+ "step": 2450
1479
+ },
1480
+ {
1481
+ "epoch": 0.79,
1482
+ "learning_rate": 1.5145656750493369e-06,
1483
+ "loss": 0.549,
1484
+ "step": 2460
1485
+ },
1486
+ {
1487
+ "epoch": 0.79,
1488
+ "learning_rate": 1.5097798439333236e-06,
1489
+ "loss": 0.5442,
1490
+ "step": 2470
1491
+ },
1492
+ {
1493
+ "epoch": 0.8,
1494
+ "learning_rate": 1.5049781846315463e-06,
1495
+ "loss": 0.5272,
1496
+ "step": 2480
1497
+ },
1498
+ {
1499
+ "epoch": 0.8,
1500
+ "learning_rate": 1.500160846231021e-06,
1501
+ "loss": 0.5324,
1502
+ "step": 2490
1503
+ },
1504
+ {
1505
+ "epoch": 0.8,
1506
+ "learning_rate": 1.4953279783055838e-06,
1507
+ "loss": 0.5234,
1508
+ "step": 2500
1509
+ },
1510
+ {
1511
+ "epoch": 0.8,
1512
+ "learning_rate": 1.490479730911249e-06,
1513
+ "loss": 0.5714,
1514
+ "step": 2510
1515
+ },
1516
+ {
1517
+ "epoch": 0.81,
1518
+ "learning_rate": 1.4856162545815477e-06,
1519
+ "loss": 0.523,
1520
+ "step": 2520
1521
+ },
1522
+ {
1523
+ "epoch": 0.81,
1524
+ "learning_rate": 1.4807377003228568e-06,
1525
+ "loss": 0.5223,
1526
+ "step": 2530
1527
+ },
1528
+ {
1529
+ "epoch": 0.81,
1530
+ "learning_rate": 1.4758442196097071e-06,
1531
+ "loss": 0.5329,
1532
+ "step": 2540
1533
+ },
1534
+ {
1535
+ "epoch": 0.82,
1536
+ "learning_rate": 1.4709359643800834e-06,
1537
+ "loss": 0.5243,
1538
+ "step": 2550
1539
+ },
1540
+ {
1541
+ "epoch": 0.82,
1542
+ "learning_rate": 1.4660130870307036e-06,
1543
+ "loss": 0.5073,
1544
+ "step": 2560
1545
+ },
1546
+ {
1547
+ "epoch": 0.82,
1548
+ "learning_rate": 1.4610757404122905e-06,
1549
+ "loss": 0.5108,
1550
+ "step": 2570
1551
+ },
1552
+ {
1553
+ "epoch": 0.83,
1554
+ "learning_rate": 1.4561240778248228e-06,
1555
+ "loss": 0.5287,
1556
+ "step": 2580
1557
+ },
1558
+ {
1559
+ "epoch": 0.83,
1560
+ "learning_rate": 1.4511582530127771e-06,
1561
+ "loss": 0.5575,
1562
+ "step": 2590
1563
+ },
1564
+ {
1565
+ "epoch": 0.83,
1566
+ "learning_rate": 1.4466770294041548e-06,
1567
+ "loss": 0.5339,
1568
+ "step": 2600
1569
+ },
1570
+ {
1571
+ "epoch": 0.84,
1572
+ "learning_rate": 1.4416847215024572e-06,
1573
+ "loss": 0.5336,
1574
+ "step": 2610
1575
+ },
1576
+ {
1577
+ "epoch": 0.84,
1578
+ "learning_rate": 1.4366786997046453e-06,
1579
+ "loss": 0.5224,
1580
+ "step": 2620
1581
+ },
1582
+ {
1583
+ "epoch": 0.84,
1584
+ "learning_rate": 1.4316591194429983e-06,
1585
+ "loss": 0.5284,
1586
+ "step": 2630
1587
+ },
1588
+ {
1589
+ "epoch": 0.85,
1590
+ "learning_rate": 1.426626136570772e-06,
1591
+ "loss": 0.5269,
1592
+ "step": 2640
1593
+ },
1594
+ {
1595
+ "epoch": 0.85,
1596
+ "learning_rate": 1.4220851219018589e-06,
1597
+ "loss": 0.5295,
1598
+ "step": 2650
1599
+ },
1600
+ {
1601
+ "epoch": 0.85,
1602
+ "learning_rate": 1.4170271049317458e-06,
1603
+ "loss": 0.5314,
1604
+ "step": 2660
1605
+ },
1606
+ {
1607
+ "epoch": 0.86,
1608
+ "learning_rate": 1.4119561396613733e-06,
1609
+ "loss": 0.5217,
1610
+ "step": 2670
1611
+ },
1612
+ {
1613
+ "epoch": 0.86,
1614
+ "learning_rate": 1.4068723835394544e-06,
1615
+ "loss": 0.5304,
1616
+ "step": 2680
1617
+ },
1618
+ {
1619
+ "epoch": 0.86,
1620
+ "learning_rate": 1.4017759944118463e-06,
1621
+ "loss": 0.5403,
1622
+ "step": 2690
1623
+ },
1624
+ {
1625
+ "epoch": 0.87,
1626
+ "learning_rate": 1.3966671305166487e-06,
1627
+ "loss": 0.5166,
1628
+ "step": 2700
1629
+ },
1630
+ {
1631
+ "epoch": 0.87,
1632
+ "learning_rate": 1.391545950479291e-06,
1633
+ "loss": 0.5413,
1634
+ "step": 2710
1635
+ },
1636
+ {
1637
+ "epoch": 0.87,
1638
+ "learning_rate": 1.3864126133076077e-06,
1639
+ "loss": 0.5279,
1640
+ "step": 2720
1641
+ },
1642
+ {
1643
+ "epoch": 0.88,
1644
+ "learning_rate": 1.3812672783869008e-06,
1645
+ "loss": 0.5086,
1646
+ "step": 2730
1647
+ },
1648
+ {
1649
+ "epoch": 0.88,
1650
+ "learning_rate": 1.376110105474991e-06,
1651
+ "loss": 0.5144,
1652
+ "step": 2740
1653
+ },
1654
+ {
1655
+ "epoch": 0.88,
1656
+ "learning_rate": 1.3709412546972574e-06,
1657
+ "loss": 0.5129,
1658
+ "step": 2750
1659
+ },
1660
+ {
1661
+ "epoch": 0.88,
1662
+ "learning_rate": 1.3657608865416665e-06,
1663
+ "loss": 0.5185,
1664
+ "step": 2760
1665
+ },
1666
+ {
1667
+ "epoch": 0.89,
1668
+ "learning_rate": 1.3605691618537876e-06,
1669
+ "loss": 0.5323,
1670
+ "step": 2770
1671
+ },
1672
+ {
1673
+ "epoch": 0.89,
1674
+ "learning_rate": 1.3553662418318008e-06,
1675
+ "loss": 0.5266,
1676
+ "step": 2780
1677
+ },
1678
+ {
1679
+ "epoch": 0.89,
1680
+ "learning_rate": 1.35015228802149e-06,
1681
+ "loss": 0.5372,
1682
+ "step": 2790
1683
+ },
1684
+ {
1685
+ "epoch": 0.9,
1686
+ "learning_rate": 1.3449274623112278e-06,
1687
+ "loss": 0.5124,
1688
+ "step": 2800
1689
+ },
1690
+ {
1691
+ "epoch": 0.9,
1692
+ "learning_rate": 1.3396919269269494e-06,
1693
+ "loss": 0.5253,
1694
+ "step": 2810
1695
+ },
1696
+ {
1697
+ "epoch": 0.9,
1698
+ "learning_rate": 1.3344458444271159e-06,
1699
+ "loss": 0.5253,
1700
+ "step": 2820
1701
+ },
1702
+ {
1703
+ "epoch": 0.91,
1704
+ "learning_rate": 1.3291893776976655e-06,
1705
+ "loss": 0.5158,
1706
+ "step": 2830
1707
+ },
1708
+ {
1709
+ "epoch": 0.91,
1710
+ "learning_rate": 1.3239226899469572e-06,
1711
+ "loss": 0.5208,
1712
+ "step": 2840
1713
+ },
1714
+ {
1715
+ "epoch": 0.91,
1716
+ "learning_rate": 1.3186459447007035e-06,
1717
+ "loss": 0.5293,
1718
+ "step": 2850
1719
+ },
1720
+ {
1721
+ "epoch": 0.92,
1722
+ "learning_rate": 1.313359305796892e-06,
1723
+ "loss": 0.5342,
1724
+ "step": 2860
1725
+ },
1726
+ {
1727
+ "epoch": 0.92,
1728
+ "learning_rate": 1.3080629373807002e-06,
1729
+ "loss": 0.5184,
1730
+ "step": 2870
1731
+ },
1732
+ {
1733
+ "epoch": 0.92,
1734
+ "learning_rate": 1.3027570038993972e-06,
1735
+ "loss": 0.543,
1736
+ "step": 2880
1737
+ },
1738
+ {
1739
+ "epoch": 0.93,
1740
+ "learning_rate": 1.2974416700972377e-06,
1741
+ "loss": 0.5352,
1742
+ "step": 2890
1743
+ },
1744
+ {
1745
+ "epoch": 0.93,
1746
+ "learning_rate": 1.2921171010103486e-06,
1747
+ "loss": 0.51,
1748
+ "step": 2900
1749
+ },
1750
+ {
1751
+ "epoch": 0.93,
1752
+ "learning_rate": 1.2867834619616036e-06,
1753
+ "loss": 0.5405,
1754
+ "step": 2910
1755
+ },
1756
+ {
1757
+ "epoch": 0.94,
1758
+ "learning_rate": 1.2814409185554894e-06,
1759
+ "loss": 0.5379,
1760
+ "step": 2920
1761
+ },
1762
+ {
1763
+ "epoch": 0.94,
1764
+ "learning_rate": 1.2760896366729656e-06,
1765
+ "loss": 0.5133,
1766
+ "step": 2930
1767
+ },
1768
+ {
1769
+ "epoch": 0.94,
1770
+ "learning_rate": 1.2707297824663124e-06,
1771
+ "loss": 0.5352,
1772
+ "step": 2940
1773
+ },
1774
+ {
1775
+ "epoch": 0.95,
1776
+ "learning_rate": 1.2653615223539738e-06,
1777
+ "loss": 0.5147,
1778
+ "step": 2950
1779
+ },
1780
+ {
1781
+ "epoch": 0.95,
1782
+ "learning_rate": 1.2599850230153876e-06,
1783
+ "loss": 0.5145,
1784
+ "step": 2960
1785
+ },
1786
+ {
1787
+ "epoch": 0.95,
1788
+ "learning_rate": 1.2546004513858138e-06,
1789
+ "loss": 0.5347,
1790
+ "step": 2970
1791
+ },
1792
+ {
1793
+ "epoch": 0.96,
1794
+ "learning_rate": 1.249207974651148e-06,
1795
+ "loss": 0.5222,
1796
+ "step": 2980
1797
+ },
1798
+ {
1799
+ "epoch": 0.96,
1800
+ "learning_rate": 1.2438077602427326e-06,
1801
+ "loss": 0.5518,
1802
+ "step": 2990
1803
+ },
1804
+ {
1805
+ "epoch": 0.96,
1806
+ "learning_rate": 1.238399975832157e-06,
1807
+ "loss": 0.5474,
1808
+ "step": 3000
1809
+ },
1810
+ {
1811
+ "epoch": 0.96,
1812
+ "learning_rate": 1.2329847893260523e-06,
1813
+ "loss": 0.5284,
1814
+ "step": 3010
1815
+ },
1816
+ {
1817
+ "epoch": 0.97,
1818
+ "learning_rate": 1.2275623688608783e-06,
1819
+ "loss": 0.5303,
1820
+ "step": 3020
1821
+ },
1822
+ {
1823
+ "epoch": 0.97,
1824
+ "learning_rate": 1.222132882797701e-06,
1825
+ "loss": 0.5255,
1826
+ "step": 3030
1827
+ },
1828
+ {
1829
+ "epoch": 0.97,
1830
+ "learning_rate": 1.2166964997169685e-06,
1831
+ "loss": 0.5114,
1832
+ "step": 3040
1833
+ },
1834
+ {
1835
+ "epoch": 0.98,
1836
+ "learning_rate": 1.2112533884132734e-06,
1837
+ "loss": 0.5186,
1838
+ "step": 3050
1839
+ },
1840
+ {
1841
+ "epoch": 0.98,
1842
+ "learning_rate": 1.2058037178901135e-06,
1843
+ "loss": 0.5251,
1844
+ "step": 3060
1845
+ },
1846
+ {
1847
+ "epoch": 0.98,
1848
+ "learning_rate": 1.2003476573546446e-06,
1849
+ "loss": 0.5334,
1850
+ "step": 3070
1851
+ },
1852
+ {
1853
+ "epoch": 0.99,
1854
+ "learning_rate": 1.194885376212426e-06,
1855
+ "loss": 0.5552,
1856
+ "step": 3080
1857
+ },
1858
+ {
1859
+ "epoch": 0.99,
1860
+ "learning_rate": 1.1894170440621618e-06,
1861
+ "loss": 0.5399,
1862
+ "step": 3090
1863
+ },
1864
+ {
1865
+ "epoch": 0.99,
1866
+ "learning_rate": 1.1839428306904332e-06,
1867
+ "loss": 0.53,
1868
+ "step": 3100
1869
+ },
1870
+ {
1871
+ "epoch": 1.0,
1872
+ "learning_rate": 1.1784629060664283e-06,
1873
+ "loss": 0.5026,
1874
+ "step": 3110
1875
+ },
1876
+ {
1877
+ "epoch": 1.0,
1878
+ "learning_rate": 1.172977440336664e-06,
1879
+ "loss": 0.5273,
1880
+ "step": 3120
1881
+ },
1882
+ {
1883
+ "epoch": 1.0,
1884
+ "learning_rate": 1.167486603819704e-06,
1885
+ "loss": 0.5394,
1886
+ "step": 3130
1887
+ },
1888
+ {
1889
+ "epoch": 1.01,
1890
+ "learning_rate": 1.1619905670008687e-06,
1891
+ "loss": 0.5064,
1892
+ "step": 3140
1893
+ },
1894
+ {
1895
+ "epoch": 1.01,
1896
+ "learning_rate": 1.1564895005269444e-06,
1897
+ "loss": 0.5017,
1898
+ "step": 3150
1899
+ },
1900
+ {
1901
+ "epoch": 1.01,
1902
+ "learning_rate": 1.1509835752008826e-06,
1903
+ "loss": 0.4899,
1904
+ "step": 3160
1905
+ },
1906
+ {
1907
+ "epoch": 1.02,
1908
+ "learning_rate": 1.1454729619764979e-06,
1909
+ "loss": 0.5225,
1910
+ "step": 3170
1911
+ },
1912
+ {
1913
+ "epoch": 1.02,
1914
+ "learning_rate": 1.1399578319531589e-06,
1915
+ "loss": 0.5183,
1916
+ "step": 3180
1917
+ },
1918
+ {
1919
+ "epoch": 1.02,
1920
+ "learning_rate": 1.1344383563704783e-06,
1921
+ "loss": 0.5085,
1922
+ "step": 3190
1923
+ },
1924
+ {
1925
+ "epoch": 1.03,
1926
+ "learning_rate": 1.1289147066029923e-06,
1927
+ "loss": 0.5064,
1928
+ "step": 3200
1929
+ },
1930
+ {
1931
+ "epoch": 1.03,
1932
+ "learning_rate": 1.1233870541548433e-06,
1933
+ "loss": 0.5008,
1934
+ "step": 3210
1935
+ },
1936
+ {
1937
+ "epoch": 1.03,
1938
+ "learning_rate": 1.1178555706544523e-06,
1939
+ "loss": 0.5301,
1940
+ "step": 3220
1941
+ },
1942
+ {
1943
+ "epoch": 1.04,
1944
+ "learning_rate": 1.1123204278491916e-06,
1945
+ "loss": 0.5334,
1946
+ "step": 3230
1947
+ },
1948
+ {
1949
+ "epoch": 1.04,
1950
+ "learning_rate": 1.1067817976000503e-06,
1951
+ "loss": 0.5323,
1952
+ "step": 3240
1953
+ },
1954
+ {
1955
+ "epoch": 1.04,
1956
+ "learning_rate": 1.1012398518763e-06,
1957
+ "loss": 0.4997,
1958
+ "step": 3250
1959
+ },
1960
+ {
1961
+ "epoch": 1.05,
1962
+ "learning_rate": 1.0956947627501553e-06,
1963
+ "loss": 0.5,
1964
+ "step": 3260
1965
+ },
1966
+ {
1967
+ "epoch": 1.05,
1968
+ "learning_rate": 1.0901467023914293e-06,
1969
+ "loss": 0.5257,
1970
+ "step": 3270
1971
+ },
1972
+ {
1973
+ "epoch": 1.05,
1974
+ "learning_rate": 1.0845958430621904e-06,
1975
+ "loss": 0.5226,
1976
+ "step": 3280
1977
+ },
1978
+ {
1979
+ "epoch": 1.05,
1980
+ "learning_rate": 1.0790423571114113e-06,
1981
+ "loss": 0.5253,
1982
+ "step": 3290
1983
+ },
1984
+ {
1985
+ "epoch": 1.06,
1986
+ "learning_rate": 1.0734864169696192e-06,
1987
+ "loss": 0.5202,
1988
+ "step": 3300
1989
+ },
1990
+ {
1991
+ "epoch": 1.06,
1992
+ "learning_rate": 1.0679281951435417e-06,
1993
+ "loss": 0.5151,
1994
+ "step": 3310
1995
+ },
1996
+ {
1997
+ "epoch": 1.06,
1998
+ "learning_rate": 1.0623678642107506e-06,
1999
+ "loss": 0.5196,
2000
+ "step": 3320
2001
+ },
2002
+ {
2003
+ "epoch": 1.07,
2004
+ "learning_rate": 1.0568055968143036e-06,
2005
+ "loss": 0.5205,
2006
+ "step": 3330
2007
+ },
2008
+ {
2009
+ "epoch": 1.07,
2010
+ "learning_rate": 1.0512415656573828e-06,
2011
+ "loss": 0.4953,
2012
+ "step": 3340
2013
+ },
2014
+ {
2015
+ "epoch": 1.07,
2016
+ "learning_rate": 1.0456759434979352e-06,
2017
+ "loss": 0.4759,
2018
+ "step": 3350
2019
+ },
2020
+ {
2021
+ "epoch": 1.08,
2022
+ "learning_rate": 1.0401089031433055e-06,
2023
+ "loss": 0.5406,
2024
+ "step": 3360
2025
+ },
2026
+ {
2027
+ "epoch": 1.08,
2028
+ "learning_rate": 1.034540617444872e-06,
2029
+ "loss": 0.4961,
2030
+ "step": 3370
2031
+ },
2032
+ {
2033
+ "epoch": 1.08,
2034
+ "learning_rate": 1.0289712592926808e-06,
2035
+ "loss": 0.5322,
2036
+ "step": 3380
2037
+ },
2038
+ {
2039
+ "epoch": 1.09,
2040
+ "learning_rate": 1.023401001610076e-06,
2041
+ "loss": 0.522,
2042
+ "step": 3390
2043
+ },
2044
+ {
2045
+ "epoch": 1.09,
2046
+ "learning_rate": 1.0178300173483312e-06,
2047
+ "loss": 0.5249,
2048
+ "step": 3400
2049
+ },
2050
+ {
2051
+ "epoch": 1.09,
2052
+ "learning_rate": 1.01225847948128e-06,
2053
+ "loss": 0.528,
2054
+ "step": 3410
2055
+ },
2056
+ {
2057
+ "epoch": 1.1,
2058
+ "learning_rate": 1.0066865609999446e-06,
2059
+ "loss": 0.5108,
2060
+ "step": 3420
2061
+ },
2062
+ {
2063
+ "epoch": 1.1,
2064
+ "learning_rate": 1.001114434907165e-06,
2065
+ "loss": 0.4848,
2066
+ "step": 3430
2067
+ },
2068
+ {
2069
+ "epoch": 1.1,
2070
+ "learning_rate": 9.955422742122276e-07,
2071
+ "loss": 0.5241,
2072
+ "step": 3440
2073
+ },
2074
+ {
2075
+ "epoch": 1.11,
2076
+ "learning_rate": 9.899702519254922e-07,
2077
+ "loss": 0.5059,
2078
+ "step": 3450
2079
+ },
2080
+ {
2081
+ "epoch": 1.11,
2082
+ "learning_rate": 9.843985410530223e-07,
2083
+ "loss": 0.5002,
2084
+ "step": 3460
2085
+ },
2086
+ {
2087
+ "epoch": 1.11,
2088
+ "learning_rate": 9.788273145912125e-07,
2089
+ "loss": 0.5022,
2090
+ "step": 3470
2091
+ },
2092
+ {
2093
+ "epoch": 1.12,
2094
+ "learning_rate": 9.73256745521415e-07,
2095
+ "loss": 0.5063,
2096
+ "step": 3480
2097
+ },
2098
+ {
2099
+ "epoch": 1.12,
2100
+ "learning_rate": 9.676870068045726e-07,
2101
+ "loss": 0.5293,
2102
+ "step": 3490
2103
+ },
2104
+ {
2105
+ "epoch": 1.12,
2106
+ "learning_rate": 9.621182713758458e-07,
2107
+ "loss": 0.5186,
2108
+ "step": 3500
2109
+ },
2110
+ {
2111
+ "epoch": 1.13,
2112
+ "learning_rate": 9.565507121392435e-07,
2113
+ "loss": 0.5322,
2114
+ "step": 3510
2115
+ },
2116
+ {
2117
+ "epoch": 1.13,
2118
+ "learning_rate": 9.509845019622556e-07,
2119
+ "loss": 0.5253,
2120
+ "step": 3520
2121
+ },
2122
+ {
2123
+ "epoch": 1.13,
2124
+ "learning_rate": 9.45419813670484e-07,
2125
+ "loss": 0.5173,
2126
+ "step": 3530
2127
+ },
2128
+ {
2129
+ "epoch": 1.13,
2130
+ "learning_rate": 9.398568200422788e-07,
2131
+ "loss": 0.5187,
2132
+ "step": 3540
2133
+ },
2134
+ {
2135
+ "epoch": 1.14,
2136
+ "learning_rate": 9.34295693803371e-07,
2137
+ "loss": 0.5233,
2138
+ "step": 3550
2139
+ },
2140
+ {
2141
+ "epoch": 1.14,
2142
+ "learning_rate": 9.287366076215121e-07,
2143
+ "loss": 0.5182,
2144
+ "step": 3560
2145
+ },
2146
+ {
2147
+ "epoch": 1.14,
2148
+ "learning_rate": 9.231797341011111e-07,
2149
+ "loss": 0.5295,
2150
+ "step": 3570
2151
+ },
2152
+ {
2153
+ "epoch": 1.15,
2154
+ "learning_rate": 9.176252457778768e-07,
2155
+ "loss": 0.5181,
2156
+ "step": 3580
2157
+ },
2158
+ {
2159
+ "epoch": 1.15,
2160
+ "learning_rate": 9.120733151134578e-07,
2161
+ "loss": 0.5262,
2162
+ "step": 3590
2163
+ },
2164
+ {
2165
+ "epoch": 1.15,
2166
+ "learning_rate": 9.065241144900921e-07,
2167
+ "loss": 0.5431,
2168
+ "step": 3600
2169
+ },
2170
+ {
2171
+ "epoch": 1.16,
2172
+ "learning_rate": 9.009778162052516e-07,
2173
+ "loss": 0.4948,
2174
+ "step": 3610
2175
+ },
2176
+ {
2177
+ "epoch": 1.16,
2178
+ "learning_rate": 8.954345924662926e-07,
2179
+ "loss": 0.517,
2180
+ "step": 3620
2181
+ },
2182
+ {
2183
+ "epoch": 1.16,
2184
+ "learning_rate": 8.898946153851106e-07,
2185
+ "loss": 0.5251,
2186
+ "step": 3630
2187
+ },
2188
+ {
2189
+ "epoch": 1.17,
2190
+ "learning_rate": 8.843580569727952e-07,
2191
+ "loss": 0.5089,
2192
+ "step": 3640
2193
+ },
2194
+ {
2195
+ "epoch": 1.17,
2196
+ "learning_rate": 8.788250891342897e-07,
2197
+ "loss": 0.5261,
2198
+ "step": 3650
2199
+ },
2200
+ {
2201
+ "epoch": 1.17,
2202
+ "learning_rate": 8.732958836630528e-07,
2203
+ "loss": 0.4978,
2204
+ "step": 3660
2205
+ },
2206
+ {
2207
+ "epoch": 1.18,
2208
+ "learning_rate": 8.677706122357257e-07,
2209
+ "loss": 0.4982,
2210
+ "step": 3670
2211
+ },
2212
+ {
2213
+ "epoch": 1.18,
2214
+ "learning_rate": 8.622494464068014e-07,
2215
+ "loss": 0.5143,
2216
+ "step": 3680
2217
+ },
2218
+ {
2219
+ "epoch": 1.18,
2220
+ "learning_rate": 8.567325576032973e-07,
2221
+ "loss": 0.4988,
2222
+ "step": 3690
2223
+ },
2224
+ {
2225
+ "epoch": 1.19,
2226
+ "learning_rate": 8.512201171194338e-07,
2227
+ "loss": 0.4856,
2228
+ "step": 3700
2229
+ },
2230
+ {
2231
+ "epoch": 1.19,
2232
+ "learning_rate": 8.457122961113148e-07,
2233
+ "loss": 0.4995,
2234
+ "step": 3710
2235
+ },
2236
+ {
2237
+ "epoch": 1.19,
2238
+ "learning_rate": 8.402092655916145e-07,
2239
+ "loss": 0.5174,
2240
+ "step": 3720
2241
+ },
2242
+ {
2243
+ "epoch": 1.2,
2244
+ "learning_rate": 8.347111964242652e-07,
2245
+ "loss": 0.5359,
2246
+ "step": 3730
2247
+ },
2248
+ {
2249
+ "epoch": 1.2,
2250
+ "learning_rate": 8.292182593191559e-07,
2251
+ "loss": 0.5072,
2252
+ "step": 3740
2253
+ },
2254
+ {
2255
+ "epoch": 1.2,
2256
+ "learning_rate": 8.237306248268292e-07,
2257
+ "loss": 0.5049,
2258
+ "step": 3750
2259
+ },
2260
+ {
2261
+ "epoch": 1.21,
2262
+ "learning_rate": 8.182484633331854e-07,
2263
+ "loss": 0.5217,
2264
+ "step": 3760
2265
+ },
2266
+ {
2267
+ "epoch": 1.21,
2268
+ "learning_rate": 8.127719450541949e-07,
2269
+ "loss": 0.5176,
2270
+ "step": 3770
2271
+ },
2272
+ {
2273
+ "epoch": 1.21,
2274
+ "learning_rate": 8.073012400306107e-07,
2275
+ "loss": 0.49,
2276
+ "step": 3780
2277
+ },
2278
+ {
2279
+ "epoch": 1.22,
2280
+ "learning_rate": 8.018365181226905e-07,
2281
+ "loss": 0.5044,
2282
+ "step": 3790
2283
+ },
2284
+ {
2285
+ "epoch": 1.22,
2286
+ "learning_rate": 7.963779490049206e-07,
2287
+ "loss": 0.5039,
2288
+ "step": 3800
2289
+ },
2290
+ {
2291
+ "epoch": 1.22,
2292
+ "learning_rate": 7.909257021607501e-07,
2293
+ "loss": 0.5094,
2294
+ "step": 3810
2295
+ },
2296
+ {
2297
+ "epoch": 1.22,
2298
+ "learning_rate": 7.854799468773269e-07,
2299
+ "loss": 0.5336,
2300
+ "step": 3820
2301
+ },
2302
+ {
2303
+ "epoch": 1.23,
2304
+ "learning_rate": 7.800408522402417e-07,
2305
+ "loss": 0.5073,
2306
+ "step": 3830
2307
+ },
2308
+ {
2309
+ "epoch": 1.23,
2310
+ "learning_rate": 7.746085871282792e-07,
2311
+ "loss": 0.5266,
2312
+ "step": 3840
2313
+ },
2314
+ {
2315
+ "epoch": 1.23,
2316
+ "learning_rate": 7.691833202081737e-07,
2317
+ "loss": 0.5199,
2318
+ "step": 3850
2319
+ },
2320
+ {
2321
+ "epoch": 1.24,
2322
+ "learning_rate": 7.637652199293712e-07,
2323
+ "loss": 0.474,
2324
+ "step": 3860
2325
+ },
2326
+ {
2327
+ "epoch": 1.24,
2328
+ "learning_rate": 7.583544545188009e-07,
2329
+ "loss": 0.5348,
2330
+ "step": 3870
2331
+ },
2332
+ {
2333
+ "epoch": 1.24,
2334
+ "learning_rate": 7.529511919756514e-07,
2335
+ "loss": 0.4942,
2336
+ "step": 3880
2337
+ },
2338
+ {
2339
+ "epoch": 1.25,
2340
+ "learning_rate": 7.475556000661546e-07,
2341
+ "loss": 0.4997,
2342
+ "step": 3890
2343
+ },
2344
+ {
2345
+ "epoch": 1.25,
2346
+ "learning_rate": 7.421678463183749e-07,
2347
+ "loss": 0.506,
2348
+ "step": 3900
2349
+ },
2350
+ {
2351
+ "epoch": 1.25,
2352
+ "learning_rate": 7.36788098017011e-07,
2353
+ "loss": 0.5053,
2354
+ "step": 3910
2355
+ },
2356
+ {
2357
+ "epoch": 1.26,
2358
+ "learning_rate": 7.314165221981987e-07,
2359
+ "loss": 0.5237,
2360
+ "step": 3920
2361
+ },
2362
+ {
2363
+ "epoch": 1.26,
2364
+ "learning_rate": 7.260532856443265e-07,
2365
+ "loss": 0.5333,
2366
+ "step": 3930
2367
+ },
2368
+ {
2369
+ "epoch": 1.26,
2370
+ "learning_rate": 7.206985548788565e-07,
2371
+ "loss": 0.5091,
2372
+ "step": 3940
2373
+ },
2374
+ {
2375
+ "epoch": 1.27,
2376
+ "learning_rate": 7.153524961611539e-07,
2377
+ "loss": 0.5137,
2378
+ "step": 3950
2379
+ },
2380
+ {
2381
+ "epoch": 1.27,
2382
+ "learning_rate": 7.100152754813251e-07,
2383
+ "loss": 0.5018,
2384
+ "step": 3960
2385
+ },
2386
+ {
2387
+ "epoch": 1.27,
2388
+ "learning_rate": 7.046870585550633e-07,
2389
+ "loss": 0.5225,
2390
+ "step": 3970
2391
+ },
2392
+ {
2393
+ "epoch": 1.28,
2394
+ "learning_rate": 6.993680108185041e-07,
2395
+ "loss": 0.4929,
2396
+ "step": 3980
2397
+ },
2398
+ {
2399
+ "epoch": 1.28,
2400
+ "learning_rate": 6.940582974230884e-07,
2401
+ "loss": 0.4791,
2402
+ "step": 3990
2403
+ },
2404
+ {
2405
+ "epoch": 1.28,
2406
+ "learning_rate": 6.88758083230434e-07,
2407
+ "loss": 0.5104,
2408
+ "step": 4000
2409
+ },
2410
+ {
2411
+ "epoch": 1.29,
2412
+ "learning_rate": 6.834675328072179e-07,
2413
+ "loss": 0.5057,
2414
+ "step": 4010
2415
+ },
2416
+ {
2417
+ "epoch": 1.29,
2418
+ "learning_rate": 6.781868104200659e-07,
2419
+ "loss": 0.528,
2420
+ "step": 4020
2421
+ },
2422
+ {
2423
+ "epoch": 1.29,
2424
+ "learning_rate": 6.72916080030453e-07,
2425
+ "loss": 0.4958,
2426
+ "step": 4030
2427
+ },
2428
+ {
2429
+ "epoch": 1.3,
2430
+ "learning_rate": 6.676555052896107e-07,
2431
+ "loss": 0.5154,
2432
+ "step": 4040
2433
+ },
2434
+ {
2435
+ "epoch": 1.3,
2436
+ "learning_rate": 6.624052495334486e-07,
2437
+ "loss": 0.5279,
2438
+ "step": 4050
2439
+ },
2440
+ {
2441
+ "epoch": 1.3,
2442
+ "learning_rate": 6.571654757774811e-07,
2443
+ "loss": 0.5204,
2444
+ "step": 4060
2445
+ },
2446
+ {
2447
+ "epoch": 1.3,
2448
+ "learning_rate": 6.51936346711766e-07,
2449
+ "loss": 0.5288,
2450
+ "step": 4070
2451
+ },
2452
+ {
2453
+ "epoch": 1.31,
2454
+ "learning_rate": 6.467180246958535e-07,
2455
+ "loss": 0.5281,
2456
+ "step": 4080
2457
+ },
2458
+ {
2459
+ "epoch": 1.31,
2460
+ "learning_rate": 6.415106717537452e-07,
2461
+ "loss": 0.5227,
2462
+ "step": 4090
2463
+ },
2464
+ {
2465
+ "epoch": 1.31,
2466
+ "learning_rate": 6.363144495688639e-07,
2467
+ "loss": 0.5171,
2468
+ "step": 4100
2469
+ },
2470
+ {
2471
+ "epoch": 1.32,
2472
+ "learning_rate": 6.311295194790311e-07,
2473
+ "loss": 0.5023,
2474
+ "step": 4110
2475
+ },
2476
+ {
2477
+ "epoch": 1.32,
2478
+ "learning_rate": 6.259560424714611e-07,
2479
+ "loss": 0.5009,
2480
+ "step": 4120
2481
+ },
2482
+ {
2483
+ "epoch": 1.32,
2484
+ "learning_rate": 6.207941791777602e-07,
2485
+ "loss": 0.5284,
2486
+ "step": 4130
2487
+ },
2488
+ {
2489
+ "epoch": 1.33,
2490
+ "learning_rate": 6.156440898689391e-07,
2491
+ "loss": 0.4981,
2492
+ "step": 4140
2493
+ },
2494
+ {
2495
+ "epoch": 1.33,
2496
+ "learning_rate": 6.10505934450438e-07,
2497
+ "loss": 0.4968,
2498
+ "step": 4150
2499
+ },
2500
+ {
2501
+ "epoch": 1.33,
2502
+ "learning_rate": 6.053798724571613e-07,
2503
+ "loss": 0.5002,
2504
+ "step": 4160
2505
+ },
2506
+ {
2507
+ "epoch": 1.34,
2508
+ "learning_rate": 6.002660630485229e-07,
2509
+ "loss": 0.5072,
2510
+ "step": 4170
2511
+ },
2512
+ {
2513
+ "epoch": 1.34,
2514
+ "learning_rate": 5.951646650035063e-07,
2515
+ "loss": 0.4958,
2516
+ "step": 4180
2517
+ },
2518
+ {
2519
+ "epoch": 1.34,
2520
+ "learning_rate": 5.900758367157333e-07,
2521
+ "loss": 0.4906,
2522
+ "step": 4190
2523
+ },
2524
+ {
2525
+ "epoch": 1.35,
2526
+ "learning_rate": 5.849997361885471e-07,
2527
+ "loss": 0.5071,
2528
+ "step": 4200
2529
+ },
2530
+ {
2531
+ "epoch": 1.35,
2532
+ "learning_rate": 5.799365210301052e-07,
2533
+ "loss": 0.5118,
2534
+ "step": 4210
2535
+ },
2536
+ {
2537
+ "epoch": 1.35,
2538
+ "learning_rate": 5.748863484484873e-07,
2539
+ "loss": 0.4934,
2540
+ "step": 4220
2541
+ },
2542
+ {
2543
+ "epoch": 1.36,
2544
+ "learning_rate": 5.698493752468125e-07,
2545
+ "loss": 0.4942,
2546
+ "step": 4230
2547
+ },
2548
+ {
2549
+ "epoch": 1.36,
2550
+ "learning_rate": 5.648257578183717e-07,
2551
+ "loss": 0.5035,
2552
+ "step": 4240
2553
+ },
2554
+ {
2555
+ "epoch": 1.36,
2556
+ "learning_rate": 5.598156521417718e-07,
2557
+ "loss": 0.4959,
2558
+ "step": 4250
2559
+ },
2560
+ {
2561
+ "epoch": 1.37,
2562
+ "learning_rate": 5.548192137760929e-07,
2563
+ "loss": 0.5081,
2564
+ "step": 4260
2565
+ },
2566
+ {
2567
+ "epoch": 1.37,
2568
+ "learning_rate": 5.49836597856057e-07,
2569
+ "loss": 0.5056,
2570
+ "step": 4270
2571
+ },
2572
+ {
2573
+ "epoch": 1.37,
2574
+ "learning_rate": 5.448679590872127e-07,
2575
+ "loss": 0.5279,
2576
+ "step": 4280
2577
+ },
2578
+ {
2579
+ "epoch": 1.38,
2580
+ "learning_rate": 5.399134517411314e-07,
2581
+ "loss": 0.5104,
2582
+ "step": 4290
2583
+ },
2584
+ {
2585
+ "epoch": 1.38,
2586
+ "learning_rate": 5.349732296506164e-07,
2587
+ "loss": 0.541,
2588
+ "step": 4300
2589
+ },
2590
+ {
2591
+ "epoch": 1.38,
2592
+ "learning_rate": 5.300474462049278e-07,
2593
+ "loss": 0.4947,
2594
+ "step": 4310
2595
+ },
2596
+ {
2597
+ "epoch": 1.39,
2598
+ "learning_rate": 5.251362543450186e-07,
2599
+ "loss": 0.5006,
2600
+ "step": 4320
2601
+ },
2602
+ {
2603
+ "epoch": 1.39,
2604
+ "learning_rate": 5.20239806558788e-07,
2605
+ "loss": 0.5475,
2606
+ "step": 4330
2607
+ },
2608
+ {
2609
+ "epoch": 1.39,
2610
+ "learning_rate": 5.153582548763447e-07,
2611
+ "loss": 0.4827,
2612
+ "step": 4340
2613
+ },
2614
+ {
2615
+ "epoch": 1.39,
2616
+ "learning_rate": 5.104917508652871e-07,
2617
+ "loss": 0.496,
2618
+ "step": 4350
2619
+ },
2620
+ {
2621
+ "epoch": 1.4,
2622
+ "learning_rate": 5.056404456259975e-07,
2623
+ "loss": 0.5047,
2624
+ "step": 4360
2625
+ },
2626
+ {
2627
+ "epoch": 1.4,
2628
+ "learning_rate": 5.008044897869523e-07,
2629
+ "loss": 0.5015,
2630
+ "step": 4370
2631
+ },
2632
+ {
2633
+ "epoch": 1.4,
2634
+ "learning_rate": 4.959840335000407e-07,
2635
+ "loss": 0.5439,
2636
+ "step": 4380
2637
+ },
2638
+ {
2639
+ "epoch": 1.41,
2640
+ "learning_rate": 4.911792264359078e-07,
2641
+ "loss": 0.4873,
2642
+ "step": 4390
2643
+ },
2644
+ {
2645
+ "epoch": 1.41,
2646
+ "learning_rate": 4.863902177793031e-07,
2647
+ "loss": 0.5053,
2648
+ "step": 4400
2649
+ },
2650
+ {
2651
+ "epoch": 1.41,
2652
+ "learning_rate": 4.816171562244522e-07,
2653
+ "loss": 0.4916,
2654
+ "step": 4410
2655
+ },
2656
+ {
2657
+ "epoch": 1.42,
2658
+ "learning_rate": 4.768601899704368e-07,
2659
+ "loss": 0.5219,
2660
+ "step": 4420
2661
+ },
2662
+ {
2663
+ "epoch": 1.42,
2664
+ "learning_rate": 4.721194667165945e-07,
2665
+ "loss": 0.5217,
2666
+ "step": 4430
2667
+ },
2668
+ {
2669
+ "epoch": 1.42,
2670
+ "learning_rate": 4.673951336579348e-07,
2671
+ "loss": 0.5125,
2672
+ "step": 4440
2673
+ },
2674
+ {
2675
+ "epoch": 1.43,
2676
+ "learning_rate": 4.626873374805642e-07,
2677
+ "loss": 0.503,
2678
+ "step": 4450
2679
+ },
2680
+ {
2681
+ "epoch": 1.43,
2682
+ "learning_rate": 4.5799622435713727e-07,
2683
+ "loss": 0.4909,
2684
+ "step": 4460
2685
+ },
2686
+ {
2687
+ "epoch": 1.43,
2688
+ "learning_rate": 4.533219399423137e-07,
2689
+ "loss": 0.5168,
2690
+ "step": 4470
2691
+ },
2692
+ {
2693
+ "epoch": 1.44,
2694
+ "learning_rate": 4.4866462936823837e-07,
2695
+ "loss": 0.5233,
2696
+ "step": 4480
2697
+ },
2698
+ {
2699
+ "epoch": 1.44,
2700
+ "learning_rate": 4.4402443724003334e-07,
2701
+ "loss": 0.5158,
2702
+ "step": 4490
2703
+ },
2704
+ {
2705
+ "epoch": 1.44,
2706
+ "learning_rate": 4.394015076313107e-07,
2707
+ "loss": 0.5162,
2708
+ "step": 4500
2709
+ },
2710
+ {
2711
+ "epoch": 1.45,
2712
+ "learning_rate": 4.347959840796961e-07,
2713
+ "loss": 0.4965,
2714
+ "step": 4510
2715
+ },
2716
+ {
2717
+ "epoch": 1.45,
2718
+ "learning_rate": 4.302080095823739e-07,
2719
+ "loss": 0.475,
2720
+ "step": 4520
2721
+ },
2722
+ {
2723
+ "epoch": 1.45,
2724
+ "learning_rate": 4.256377265916463e-07,
2725
+ "loss": 0.5333,
2726
+ "step": 4530
2727
+ },
2728
+ {
2729
+ "epoch": 1.46,
2730
+ "learning_rate": 4.210852770105121e-07,
2731
+ "loss": 0.514,
2732
+ "step": 4540
2733
+ },
2734
+ {
2735
+ "epoch": 1.46,
2736
+ "learning_rate": 4.165508021882583e-07,
2737
+ "loss": 0.5091,
2738
+ "step": 4550
2739
+ },
2740
+ {
2741
+ "epoch": 1.46,
2742
+ "learning_rate": 4.120344429160726e-07,
2743
+ "loss": 0.4908,
2744
+ "step": 4560
2745
+ },
2746
+ {
2747
+ "epoch": 1.47,
2748
+ "learning_rate": 4.0753633942267263e-07,
2749
+ "loss": 0.5042,
2750
+ "step": 4570
2751
+ },
2752
+ {
2753
+ "epoch": 1.47,
2754
+ "learning_rate": 4.030566313699505e-07,
2755
+ "loss": 0.5021,
2756
+ "step": 4580
2757
+ },
2758
+ {
2759
+ "epoch": 1.47,
2760
+ "learning_rate": 3.9859545784863735e-07,
2761
+ "loss": 0.5313,
2762
+ "step": 4590
2763
+ },
2764
+ {
2765
+ "epoch": 1.47,
2766
+ "learning_rate": 3.9415295737398416e-07,
2767
+ "loss": 0.5158,
2768
+ "step": 4600
2769
+ },
2770
+ {
2771
+ "epoch": 1.48,
2772
+ "learning_rate": 3.8972926788146285e-07,
2773
+ "loss": 0.5122,
2774
+ "step": 4610
2775
+ },
2776
+ {
2777
+ "epoch": 1.48,
2778
+ "learning_rate": 3.853245267224794e-07,
2779
+ "loss": 0.5112,
2780
+ "step": 4620
2781
+ },
2782
+ {
2783
+ "epoch": 1.48,
2784
+ "learning_rate": 3.809388706601142e-07,
2785
+ "loss": 0.5023,
2786
+ "step": 4630
2787
+ },
2788
+ {
2789
+ "epoch": 1.49,
2790
+ "learning_rate": 3.7657243586487197e-07,
2791
+ "loss": 0.4939,
2792
+ "step": 4640
2793
+ },
2794
+ {
2795
+ "epoch": 1.49,
2796
+ "learning_rate": 3.7222535791045595e-07,
2797
+ "loss": 0.5016,
2798
+ "step": 4650
2799
+ },
2800
+ {
2801
+ "epoch": 1.49,
2802
+ "learning_rate": 3.6789777176955703e-07,
2803
+ "loss": 0.5089,
2804
+ "step": 4660
2805
+ },
2806
+ {
2807
+ "epoch": 1.5,
2808
+ "learning_rate": 3.6358981180966375e-07,
2809
+ "loss": 0.4797,
2810
+ "step": 4670
2811
+ },
2812
+ {
2813
+ "epoch": 1.5,
2814
+ "learning_rate": 3.593016117888901e-07,
2815
+ "loss": 0.5253,
2816
+ "step": 4680
2817
+ },
2818
+ {
2819
+ "epoch": 1.5,
2820
+ "learning_rate": 3.5503330485182217e-07,
2821
+ "loss": 0.4959,
2822
+ "step": 4690
2823
+ },
2824
+ {
2825
+ "epoch": 1.51,
2826
+ "learning_rate": 3.5120894674136217e-07,
2827
+ "loss": 0.507,
2828
+ "step": 4700
2829
+ },
2830
+ {
2831
+ "epoch": 1.51,
2832
+ "learning_rate": 3.469788012612012e-07,
2833
+ "loss": 0.4975,
2834
+ "step": 4710
2835
+ },
2836
+ {
2837
+ "epoch": 1.51,
2838
+ "learning_rate": 3.427689314764486e-07,
2839
+ "loss": 0.5109,
2840
+ "step": 4720
2841
+ },
2842
+ {
2843
+ "epoch": 1.52,
2844
+ "learning_rate": 3.385794680996115e-07,
2845
+ "loss": 0.5059,
2846
+ "step": 4730
2847
+ },
2848
+ {
2849
+ "epoch": 1.52,
2850
+ "learning_rate": 3.344105412095959e-07,
2851
+ "loss": 0.507,
2852
+ "step": 4740
2853
+ },
2854
+ {
2855
+ "epoch": 1.52,
2856
+ "learning_rate": 3.302622802476694e-07,
2857
+ "loss": 0.4812,
2858
+ "step": 4750
2859
+ },
2860
+ {
2861
+ "epoch": 1.53,
2862
+ "learning_rate": 3.26134814013442e-07,
2863
+ "loss": 0.4863,
2864
+ "step": 4760
2865
+ },
2866
+ {
2867
+ "epoch": 1.53,
2868
+ "learning_rate": 3.2202827066086734e-07,
2869
+ "loss": 0.4808,
2870
+ "step": 4770
2871
+ },
2872
+ {
2873
+ "epoch": 1.53,
2874
+ "learning_rate": 3.179427776942626e-07,
2875
+ "loss": 0.5236,
2876
+ "step": 4780
2877
+ },
2878
+ {
2879
+ "epoch": 1.54,
2880
+ "learning_rate": 3.1387846196435043e-07,
2881
+ "loss": 0.473,
2882
+ "step": 4790
2883
+ },
2884
+ {
2885
+ "epoch": 1.54,
2886
+ "learning_rate": 3.098354496643198e-07,
2887
+ "loss": 0.5148,
2888
+ "step": 4800
2889
+ },
2890
+ {
2891
+ "epoch": 1.54,
2892
+ "learning_rate": 3.0581386632590943e-07,
2893
+ "loss": 0.4956,
2894
+ "step": 4810
2895
+ },
2896
+ {
2897
+ "epoch": 1.55,
2898
+ "learning_rate": 3.0181383681550756e-07,
2899
+ "loss": 0.4898,
2900
+ "step": 4820
2901
+ },
2902
+ {
2903
+ "epoch": 1.55,
2904
+ "learning_rate": 2.9783548533027636e-07,
2905
+ "loss": 0.4821,
2906
+ "step": 4830
2907
+ },
2908
+ {
2909
+ "epoch": 1.55,
2910
+ "learning_rate": 2.9387893539429673e-07,
2911
+ "loss": 0.507,
2912
+ "step": 4840
2913
+ },
2914
+ {
2915
+ "epoch": 1.55,
2916
+ "learning_rate": 2.899443098547311e-07,
2917
+ "loss": 0.5288,
2918
+ "step": 4850
2919
+ },
2920
+ {
2921
+ "epoch": 1.56,
2922
+ "learning_rate": 2.860317308780098e-07,
2923
+ "loss": 0.5102,
2924
+ "step": 4860
2925
+ },
2926
+ {
2927
+ "epoch": 1.56,
2928
+ "learning_rate": 2.821413199460384e-07,
2929
+ "loss": 0.5139,
2930
+ "step": 4870
2931
+ },
2932
+ {
2933
+ "epoch": 1.56,
2934
+ "learning_rate": 2.782731978524263e-07,
2935
+ "loss": 0.5116,
2936
+ "step": 4880
2937
+ },
2938
+ {
2939
+ "epoch": 1.57,
2940
+ "learning_rate": 2.7442748469873334e-07,
2941
+ "loss": 0.5268,
2942
+ "step": 4890
2943
+ },
2944
+ {
2945
+ "epoch": 1.57,
2946
+ "learning_rate": 2.7060429989074486e-07,
2947
+ "loss": 0.5082,
2948
+ "step": 4900
2949
+ },
2950
+ {
2951
+ "epoch": 1.57,
2952
+ "learning_rate": 2.6680376213476074e-07,
2953
+ "loss": 0.5057,
2954
+ "step": 4910
2955
+ },
2956
+ {
2957
+ "epoch": 1.58,
2958
+ "learning_rate": 2.6302598943391285e-07,
2959
+ "loss": 0.5126,
2960
+ "step": 4920
2961
+ },
2962
+ {
2963
+ "epoch": 1.58,
2964
+ "learning_rate": 2.592710990844972e-07,
2965
+ "loss": 0.4905,
2966
+ "step": 4930
2967
+ },
2968
+ {
2969
+ "epoch": 1.58,
2970
+ "learning_rate": 2.5553920767233617e-07,
2971
+ "loss": 0.4916,
2972
+ "step": 4940
2973
+ },
2974
+ {
2975
+ "epoch": 1.59,
2976
+ "learning_rate": 2.518304310691558e-07,
2977
+ "loss": 0.4932,
2978
+ "step": 4950
2979
+ },
2980
+ {
2981
+ "epoch": 1.59,
2982
+ "learning_rate": 2.48144884428989e-07,
2983
+ "loss": 0.4999,
2984
+ "step": 4960
2985
+ },
2986
+ {
2987
+ "epoch": 1.59,
2988
+ "learning_rate": 2.444826821846007e-07,
2989
+ "loss": 0.4867,
2990
+ "step": 4970
2991
+ },
2992
+ {
2993
+ "epoch": 1.6,
2994
+ "learning_rate": 2.4084393804393345e-07,
2995
+ "loss": 0.5218,
2996
+ "step": 4980
2997
+ },
2998
+ {
2999
+ "epoch": 1.6,
3000
+ "learning_rate": 2.3722876498657906e-07,
3001
+ "loss": 0.4948,
3002
+ "step": 4990
3003
+ },
3004
+ {
3005
+ "epoch": 1.6,
3006
+ "learning_rate": 2.336372752602671e-07,
3007
+ "loss": 0.5039,
3008
+ "step": 5000
3009
+ },
3010
+ {
3011
+ "epoch": 1.61,
3012
+ "learning_rate": 2.300695803773841e-07,
3013
+ "loss": 0.4744,
3014
+ "step": 5010
3015
+ },
3016
+ {
3017
+ "epoch": 1.61,
3018
+ "learning_rate": 2.2652579111150806e-07,
3019
+ "loss": 0.4962,
3020
+ "step": 5020
3021
+ },
3022
+ {
3023
+ "epoch": 1.61,
3024
+ "learning_rate": 2.2300601749396986e-07,
3025
+ "loss": 0.5068,
3026
+ "step": 5030
3027
+ },
3028
+ {
3029
+ "epoch": 1.62,
3030
+ "learning_rate": 2.1951036881043704e-07,
3031
+ "loss": 0.5082,
3032
+ "step": 5040
3033
+ },
3034
+ {
3035
+ "epoch": 1.62,
3036
+ "learning_rate": 2.1603895359752145e-07,
3037
+ "loss": 0.4763,
3038
+ "step": 5050
3039
+ },
3040
+ {
3041
+ "epoch": 1.62,
3042
+ "learning_rate": 2.125918796394076e-07,
3043
+ "loss": 0.503,
3044
+ "step": 5060
3045
+ },
3046
+ {
3047
+ "epoch": 1.63,
3048
+ "learning_rate": 2.0916925396450692e-07,
3049
+ "loss": 0.5045,
3050
+ "step": 5070
3051
+ },
3052
+ {
3053
+ "epoch": 1.63,
3054
+ "learning_rate": 2.0577118284213447e-07,
3055
+ "loss": 0.5247,
3056
+ "step": 5080
3057
+ },
3058
+ {
3059
+ "epoch": 1.63,
3060
+ "learning_rate": 2.0239777177921025e-07,
3061
+ "loss": 0.4982,
3062
+ "step": 5090
3063
+ },
3064
+ {
3065
+ "epoch": 1.64,
3066
+ "learning_rate": 1.9904912551698193e-07,
3067
+ "loss": 0.5274,
3068
+ "step": 5100
3069
+ },
3070
+ {
3071
+ "epoch": 1.64,
3072
+ "learning_rate": 1.957253480277733e-07,
3073
+ "loss": 0.479,
3074
+ "step": 5110
3075
+ },
3076
+ {
3077
+ "epoch": 1.64,
3078
+ "learning_rate": 1.9242654251175683e-07,
3079
+ "loss": 0.4976,
3080
+ "step": 5120
3081
+ },
3082
+ {
3083
+ "epoch": 1.64,
3084
+ "learning_rate": 1.891528113937473e-07,
3085
+ "loss": 0.5086,
3086
+ "step": 5130
3087
+ },
3088
+ {
3089
+ "epoch": 1.65,
3090
+ "learning_rate": 1.8590425632002426e-07,
3091
+ "loss": 0.509,
3092
+ "step": 5140
3093
+ },
3094
+ {
3095
+ "epoch": 1.65,
3096
+ "learning_rate": 1.8268097815517402e-07,
3097
+ "loss": 0.5139,
3098
+ "step": 5150
3099
+ },
3100
+ {
3101
+ "epoch": 1.65,
3102
+ "learning_rate": 1.7948307697895937e-07,
3103
+ "loss": 0.5068,
3104
+ "step": 5160
3105
+ },
3106
+ {
3107
+ "epoch": 1.66,
3108
+ "learning_rate": 1.7631065208320995e-07,
3109
+ "loss": 0.5124,
3110
+ "step": 5170
3111
+ },
3112
+ {
3113
+ "epoch": 1.66,
3114
+ "learning_rate": 1.7316380196874236e-07,
3115
+ "loss": 0.5039,
3116
+ "step": 5180
3117
+ },
3118
+ {
3119
+ "epoch": 1.66,
3120
+ "learning_rate": 1.7004262434229922e-07,
3121
+ "loss": 0.5173,
3122
+ "step": 5190
3123
+ },
3124
+ {
3125
+ "epoch": 1.67,
3126
+ "learning_rate": 1.6694721611351703e-07,
3127
+ "loss": 0.5033,
3128
+ "step": 5200
3129
+ },
3130
+ {
3131
+ "epoch": 1.67,
3132
+ "learning_rate": 1.6387767339191615e-07,
3133
+ "loss": 0.5017,
3134
+ "step": 5210
3135
+ },
3136
+ {
3137
+ "epoch": 1.67,
3138
+ "learning_rate": 1.6083409148391803e-07,
3139
+ "loss": 0.5063,
3140
+ "step": 5220
3141
+ },
3142
+ {
3143
+ "epoch": 1.68,
3144
+ "learning_rate": 1.5781656488988447e-07,
3145
+ "loss": 0.509,
3146
+ "step": 5230
3147
+ },
3148
+ {
3149
+ "epoch": 1.68,
3150
+ "learning_rate": 1.5482518730118443e-07,
3151
+ "loss": 0.5369,
3152
+ "step": 5240
3153
+ },
3154
+ {
3155
+ "epoch": 1.68,
3156
+ "learning_rate": 1.518600515972851e-07,
3157
+ "loss": 0.5068,
3158
+ "step": 5250
3159
+ },
3160
+ {
3161
+ "epoch": 1.69,
3162
+ "learning_rate": 1.4892124984286736e-07,
3163
+ "loss": 0.5026,
3164
+ "step": 5260
3165
+ },
3166
+ {
3167
+ "epoch": 1.69,
3168
+ "learning_rate": 1.460088732849678e-07,
3169
+ "loss": 0.4871,
3170
+ "step": 5270
3171
+ },
3172
+ {
3173
+ "epoch": 1.69,
3174
+ "learning_rate": 1.4312301235014534e-07,
3175
+ "loss": 0.5198,
3176
+ "step": 5280
3177
+ },
3178
+ {
3179
+ "epoch": 1.7,
3180
+ "learning_rate": 1.4026375664167434e-07,
3181
+ "loss": 0.4999,
3182
+ "step": 5290
3183
+ },
3184
+ {
3185
+ "epoch": 1.7,
3186
+ "learning_rate": 1.3743119493676102e-07,
3187
+ "loss": 0.4855,
3188
+ "step": 5300
3189
+ },
3190
+ {
3191
+ "epoch": 1.7,
3192
+ "learning_rate": 1.3462541518378823e-07,
3193
+ "loss": 0.5142,
3194
+ "step": 5310
3195
+ },
3196
+ {
3197
+ "epoch": 1.71,
3198
+ "learning_rate": 1.318465044995841e-07,
3199
+ "loss": 0.5197,
3200
+ "step": 5320
3201
+ },
3202
+ {
3203
+ "epoch": 1.71,
3204
+ "learning_rate": 1.290945491667178e-07,
3205
+ "loss": 0.5181,
3206
+ "step": 5330
3207
+ },
3208
+ {
3209
+ "epoch": 1.71,
3210
+ "learning_rate": 1.2636963463081996e-07,
3211
+ "loss": 0.5227,
3212
+ "step": 5340
3213
+ },
3214
+ {
3215
+ "epoch": 1.72,
3216
+ "learning_rate": 1.2367184549792963e-07,
3217
+ "loss": 0.5132,
3218
+ "step": 5350
3219
+ },
3220
+ {
3221
+ "epoch": 1.72,
3222
+ "learning_rate": 1.2100126553186784e-07,
3223
+ "loss": 0.5024,
3224
+ "step": 5360
3225
+ },
3226
+ {
3227
+ "epoch": 1.72,
3228
+ "learning_rate": 1.1835797765163601e-07,
3229
+ "loss": 0.5134,
3230
+ "step": 5370
3231
+ },
3232
+ {
3233
+ "epoch": 1.72,
3234
+ "learning_rate": 1.1574206392884279e-07,
3235
+ "loss": 0.4833,
3236
+ "step": 5380
3237
+ },
3238
+ {
3239
+ "epoch": 1.73,
3240
+ "learning_rate": 1.1315360558515408e-07,
3241
+ "loss": 0.496,
3242
+ "step": 5390
3243
+ },
3244
+ {
3245
+ "epoch": 1.73,
3246
+ "learning_rate": 1.1059268298977286e-07,
3247
+ "loss": 0.5051,
3248
+ "step": 5400
3249
+ },
3250
+ {
3251
+ "epoch": 1.73,
3252
+ "learning_rate": 1.0805937565694189e-07,
3253
+ "loss": 0.5138,
3254
+ "step": 5410
3255
+ },
3256
+ {
3257
+ "epoch": 1.74,
3258
+ "learning_rate": 1.055537622434769e-07,
3259
+ "loss": 0.5038,
3260
+ "step": 5420
3261
+ },
3262
+ {
3263
+ "epoch": 1.74,
3264
+ "learning_rate": 1.0307592054632297e-07,
3265
+ "loss": 0.4939,
3266
+ "step": 5430
3267
+ },
3268
+ {
3269
+ "epoch": 1.74,
3270
+ "learning_rate": 1.0062592750013944e-07,
3271
+ "loss": 0.498,
3272
+ "step": 5440
3273
+ },
3274
+ {
3275
+ "epoch": 1.75,
3276
+ "learning_rate": 9.820385917491114e-08,
3277
+ "loss": 0.4849,
3278
+ "step": 5450
3279
+ },
3280
+ {
3281
+ "epoch": 1.75,
3282
+ "learning_rate": 9.580979077358708e-08,
3283
+ "loss": 0.5012,
3284
+ "step": 5460
3285
+ },
3286
+ {
3287
+ "epoch": 1.75,
3288
+ "learning_rate": 9.344379662974412e-08,
3289
+ "loss": 0.509,
3290
+ "step": 5470
3291
+ },
3292
+ {
3293
+ "epoch": 1.76,
3294
+ "learning_rate": 9.110595020528e-08,
3295
+ "loss": 0.5329,
3296
+ "step": 5480
3297
+ },
3298
+ {
3299
+ "epoch": 1.76,
3300
+ "learning_rate": 8.879632408813275e-08,
3301
+ "loss": 0.5232,
3302
+ "step": 5490
3303
+ },
3304
+ {
3305
+ "epoch": 1.76,
3306
+ "learning_rate": 8.65149899900256e-08,
3307
+ "loss": 0.5319,
3308
+ "step": 5500
3309
+ },
3310
+ {
3311
+ "epoch": 1.77,
3312
+ "learning_rate": 8.426201874424154e-08,
3313
+ "loss": 0.4951,
3314
+ "step": 5510
3315
+ },
3316
+ {
3317
+ "epoch": 1.77,
3318
+ "learning_rate": 8.203748030342328e-08,
3319
+ "loss": 0.5111,
3320
+ "step": 5520
3321
+ },
3322
+ {
3323
+ "epoch": 1.77,
3324
+ "learning_rate": 7.98414437374021e-08,
3325
+ "loss": 0.4982,
3326
+ "step": 5530
3327
+ },
3328
+ {
3329
+ "epoch": 1.78,
3330
+ "learning_rate": 7.767397723105252e-08,
3331
+ "loss": 0.4836,
3332
+ "step": 5540
3333
+ },
3334
+ {
3335
+ "epoch": 1.78,
3336
+ "learning_rate": 7.553514808217543e-08,
3337
+ "loss": 0.4937,
3338
+ "step": 5550
3339
+ },
3340
+ {
3341
+ "epoch": 1.78,
3342
+ "learning_rate": 7.342502269940876e-08,
3343
+ "loss": 0.4974,
3344
+ "step": 5560
3345
+ },
3346
+ {
3347
+ "epoch": 1.79,
3348
+ "learning_rate": 7.134366660016578e-08,
3349
+ "loss": 0.4974,
3350
+ "step": 5570
3351
+ },
3352
+ {
3353
+ "epoch": 1.79,
3354
+ "learning_rate": 6.929114440859973e-08,
3355
+ "loss": 0.5046,
3356
+ "step": 5580
3357
+ },
3358
+ {
3359
+ "epoch": 1.79,
3360
+ "learning_rate": 6.726751985359869e-08,
3361
+ "loss": 0.5193,
3362
+ "step": 5590
3363
+ },
3364
+ {
3365
+ "epoch": 1.8,
3366
+ "learning_rate": 6.527285576680608e-08,
3367
+ "loss": 0.5089,
3368
+ "step": 5600
3369
+ },
3370
+ {
3371
+ "epoch": 1.8,
3372
+ "learning_rate": 6.330721408067008e-08,
3373
+ "loss": 0.4935,
3374
+ "step": 5610
3375
+ },
3376
+ {
3377
+ "epoch": 1.8,
3378
+ "learning_rate": 6.137065582652034e-08,
3379
+ "loss": 0.5122,
3380
+ "step": 5620
3381
+ },
3382
+ {
3383
+ "epoch": 1.8,
3384
+ "learning_rate": 5.9463241132673313e-08,
3385
+ "loss": 0.4797,
3386
+ "step": 5630
3387
+ },
3388
+ {
3389
+ "epoch": 1.81,
3390
+ "learning_rate": 5.7585029222565254e-08,
3391
+ "loss": 0.5062,
3392
+ "step": 5640
3393
+ },
3394
+ {
3395
+ "epoch": 1.81,
3396
+ "learning_rate": 5.573607841291328e-08,
3397
+ "loss": 0.489,
3398
+ "step": 5650
3399
+ },
3400
+ {
3401
+ "epoch": 1.81,
3402
+ "learning_rate": 5.391644611190527e-08,
3403
+ "loss": 0.5044,
3404
+ "step": 5660
3405
+ },
3406
+ {
3407
+ "epoch": 1.82,
3408
+ "learning_rate": 5.2126188817416374e-08,
3409
+ "loss": 0.5254,
3410
+ "step": 5670
3411
+ },
3412
+ {
3413
+ "epoch": 1.82,
3414
+ "learning_rate": 5.036536211525555e-08,
3415
+ "loss": 0.5024,
3416
+ "step": 5680
3417
+ },
3418
+ {
3419
+ "epoch": 1.82,
3420
+ "learning_rate": 4.863402067743949e-08,
3421
+ "loss": 0.5075,
3422
+ "step": 5690
3423
+ },
3424
+ {
3425
+ "epoch": 1.83,
3426
+ "learning_rate": 4.6932218260495205e-08,
3427
+ "loss": 0.4883,
3428
+ "step": 5700
3429
+ },
3430
+ {
3431
+ "epoch": 1.83,
3432
+ "learning_rate": 4.5260007703790685e-08,
3433
+ "loss": 0.4976,
3434
+ "step": 5710
3435
+ },
3436
+ {
3437
+ "epoch": 1.83,
3438
+ "learning_rate": 4.361744092789443e-08,
3439
+ "loss": 0.4856,
3440
+ "step": 5720
3441
+ },
3442
+ {
3443
+ "epoch": 1.84,
3444
+ "learning_rate": 4.200456893296323e-08,
3445
+ "loss": 0.5119,
3446
+ "step": 5730
3447
+ },
3448
+ {
3449
+ "epoch": 1.84,
3450
+ "learning_rate": 4.042144179715945e-08,
3451
+ "loss": 0.5115,
3452
+ "step": 5740
3453
+ },
3454
+ {
3455
+ "epoch": 1.84,
3456
+ "learning_rate": 3.8868108675094605e-08,
3457
+ "loss": 0.5033,
3458
+ "step": 5750
3459
+ },
3460
+ {
3461
+ "epoch": 1.85,
3462
+ "learning_rate": 3.734461779630438e-08,
3463
+ "loss": 0.5275,
3464
+ "step": 5760
3465
+ },
3466
+ {
3467
+ "epoch": 1.85,
3468
+ "learning_rate": 3.5851016463751326e-08,
3469
+ "loss": 0.4971,
3470
+ "step": 5770
3471
+ },
3472
+ {
3473
+ "epoch": 1.85,
3474
+ "learning_rate": 3.4387351052354794e-08,
3475
+ "loss": 0.4897,
3476
+ "step": 5780
3477
+ },
3478
+ {
3479
+ "epoch": 1.86,
3480
+ "learning_rate": 3.295366700755253e-08,
3481
+ "loss": 0.5228,
3482
+ "step": 5790
3483
+ },
3484
+ {
3485
+ "epoch": 1.86,
3486
+ "learning_rate": 3.155000884388837e-08,
3487
+ "loss": 0.4984,
3488
+ "step": 5800
3489
+ },
3490
+ {
3491
+ "epoch": 1.86,
3492
+ "learning_rate": 3.017642014363153e-08,
3493
+ "loss": 0.5088,
3494
+ "step": 5810
3495
+ },
3496
+ {
3497
+ "epoch": 1.87,
3498
+ "learning_rate": 2.883294355542165e-08,
3499
+ "loss": 0.5066,
3500
+ "step": 5820
3501
+ },
3502
+ {
3503
+ "epoch": 1.87,
3504
+ "learning_rate": 2.7519620792946342e-08,
3505
+ "loss": 0.5143,
3506
+ "step": 5830
3507
+ },
3508
+ {
3509
+ "epoch": 1.87,
3510
+ "learning_rate": 2.6236492633644803e-08,
3511
+ "loss": 0.4984,
3512
+ "step": 5840
3513
+ },
3514
+ {
3515
+ "epoch": 1.88,
3516
+ "learning_rate": 2.4983598917442506e-08,
3517
+ "loss": 0.4915,
3518
+ "step": 5850
3519
+ },
3520
+ {
3521
+ "epoch": 1.88,
3522
+ "learning_rate": 2.376097854551329e-08,
3523
+ "loss": 0.5076,
3524
+ "step": 5860
3525
+ },
3526
+ {
3527
+ "epoch": 1.88,
3528
+ "learning_rate": 2.2568669479072543e-08,
3529
+ "loss": 0.4881,
3530
+ "step": 5870
3531
+ },
3532
+ {
3533
+ "epoch": 1.89,
3534
+ "learning_rate": 2.140670873819772e-08,
3535
+ "loss": 0.4999,
3536
+ "step": 5880
3537
+ },
3538
+ {
3539
+ "epoch": 1.89,
3540
+ "learning_rate": 2.0275132400679572e-08,
3541
+ "loss": 0.4984,
3542
+ "step": 5890
3543
+ },
3544
+ {
3545
+ "epoch": 1.89,
3546
+ "learning_rate": 1.9173975600901394e-08,
3547
+ "loss": 0.4792,
3548
+ "step": 5900
3549
+ },
3550
+ {
3551
+ "epoch": 1.89,
3552
+ "learning_rate": 1.8103272528748436e-08,
3553
+ "loss": 0.4904,
3554
+ "step": 5910
3555
+ },
3556
+ {
3557
+ "epoch": 1.9,
3558
+ "learning_rate": 1.7063056428546552e-08,
3559
+ "loss": 0.4989,
3560
+ "step": 5920
3561
+ },
3562
+ {
3563
+ "epoch": 1.9,
3564
+ "learning_rate": 1.6053359598029226e-08,
3565
+ "loss": 0.5054,
3566
+ "step": 5930
3567
+ },
3568
+ {
3569
+ "epoch": 1.9,
3570
+ "learning_rate": 1.50742133873355e-08,
3571
+ "loss": 0.4859,
3572
+ "step": 5940
3573
+ },
3574
+ {
3575
+ "epoch": 1.91,
3576
+ "learning_rate": 1.4125648198036633e-08,
3577
+ "loss": 0.4821,
3578
+ "step": 5950
3579
+ },
3580
+ {
3581
+ "epoch": 1.91,
3582
+ "learning_rate": 1.3207693482191307e-08,
3583
+ "loss": 0.4995,
3584
+ "step": 5960
3585
+ },
3586
+ {
3587
+ "epoch": 1.91,
3588
+ "learning_rate": 1.232037774143213e-08,
3589
+ "loss": 0.5035,
3590
+ "step": 5970
3591
+ },
3592
+ {
3593
+ "epoch": 1.92,
3594
+ "learning_rate": 1.146372852608024e-08,
3595
+ "loss": 0.5021,
3596
+ "step": 5980
3597
+ },
3598
+ {
3599
+ "epoch": 1.92,
3600
+ "learning_rate": 1.063777243428965e-08,
3601
+ "loss": 0.4814,
3602
+ "step": 5990
3603
+ },
3604
+ {
3605
+ "epoch": 1.92,
3606
+ "learning_rate": 9.842535111222238e-09,
3607
+ "loss": 0.4901,
3608
+ "step": 6000
3609
+ }
3610
+ ],
3611
+ "max_steps": 6238,
3612
+ "num_train_epochs": 2,
3613
+ "total_flos": 2.0920322636916982e+18,
3614
+ "trial_name": null,
3615
+ "trial_params": null
3616
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74b0bb469b24dc9f95aa9933cf8889e25be4c02c6f00fe7e1798280e23e0a131
3
+ size 4347