Davlan commited on
Commit
eafb3f7
1 Parent(s): f2e85a6

add MT model

Browse files
all_results.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "predict_bleu": 11.1534,
4
+ "predict_gen_len": 46.5521,
5
+ "predict_loss": 2.472553014755249,
6
+ "predict_runtime": 452.3537,
7
+ "predict_samples": 998,
8
+ "predict_samples_per_second": 2.206,
9
+ "predict_steps_per_second": 0.221,
10
+ "train_loss": 0.564786915779832,
11
+ "train_runtime": 39960.6259,
12
+ "train_samples": 663898,
13
+ "train_samples_per_second": 49.841,
14
+ "train_steps_per_second": 4.984
15
+ }
config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/m2m100_418M",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "relu",
5
+ "architectures": [
6
+ "M2M100ForConditionalGeneration"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 0,
10
+ "d_model": 1024,
11
+ "decoder_attention_heads": 16,
12
+ "decoder_ffn_dim": 4096,
13
+ "decoder_layerdrop": 0.05,
14
+ "decoder_layers": 12,
15
+ "decoder_start_token_id": 2,
16
+ "dropout": 0.1,
17
+ "early_stopping": true,
18
+ "encoder_attention_heads": 16,
19
+ "encoder_ffn_dim": 4096,
20
+ "encoder_layerdrop": 0.05,
21
+ "encoder_layers": 12,
22
+ "eos_token_id": 2,
23
+ "forced_bos_token_id": 128103,
24
+ "gradient_checkpointing": false,
25
+ "init_std": 0.02,
26
+ "is_encoder_decoder": true,
27
+ "max_length": 200,
28
+ "max_position_embeddings": 1024,
29
+ "model_type": "m2m_100",
30
+ "num_beams": 5,
31
+ "num_hidden_layers": 12,
32
+ "pad_token_id": 1,
33
+ "scale_embedding": true,
34
+ "torch_dtype": "float32",
35
+ "transformers_version": "4.10.3",
36
+ "use_cache": true,
37
+ "vocab_size": 128112
38
+ }
generated_predictions.txt ADDED
The diff for this file is too large to render. See raw diff
 
predict_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "predict_bleu": 11.1534,
3
+ "predict_gen_len": 46.5521,
4
+ "predict_loss": 2.472553014755249,
5
+ "predict_runtime": 452.3537,
6
+ "predict_samples": 998,
7
+ "predict_samples_per_second": 2.206,
8
+ "predict_steps_per_second": 0.221
9
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d26f9cf7060994bc0c30a47a1e322a3e93e0fccc3deea940e2bbbb28a356094d
3
+ size 1935792071
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d8f7c76ed2a5e0822be39f0a4f95a55eb19c78f4593ce609e2edbc2aea4d380a
3
+ size 2423393
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "additional_special_tokens": ["__af__", "__am__", "__ar__", "__ast__", "__az__", "__ba__", "__be__", "__bg__", "__bn__", "__br__", "__bs__", "__ca__", "__ceb__", "__cs__", "__cy__", "__da__", "__de__", "__el__", "__en__", "__es__", "__et__", "__fa__", "__ff__", "__fi__", "__fr__", "__fy__", "__ga__", "__gd__", "__gl__", "__gu__", "__ha__", "__he__", "__hi__", "__hr__", "__ht__", "__hu__", "__hy__", "__id__", "__ig__", "__ilo__", "__is__", "__it__", "__ja__", "__jv__", "__ka__", "__kk__", "__km__", "__kn__", "__ko__", "__lb__", "__lg__", "__ln__", "__lo__", "__lt__", "__lv__", "__mg__", "__mk__", "__ml__", "__mn__", "__mr__", "__ms__", "__my__", "__ne__", "__nl__", "__no__", "__ns__", "__oc__", "__or__", "__pa__", "__pl__", "__ps__", "__pt__", "__ro__", "__ru__", "__sd__", "__si__", "__sk__", "__sl__", "__so__", "__sq__", "__sr__", "__ss__", "__su__", "__sv__", "__sw__", "__ta__", "__th__", "__tl__", "__tn__", "__tr__", "__uk__", "__ur__", "__uz__", "__vi__", "__wo__", "__xh__", "__yi__", "__yo__", "__zh__", "__zu__"]}
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"src_lang": null, "tgt_lang": null, "bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "sp_model_kwargs": {}, "special_tokens_map_file": "m2m_100_1.2B_v2/special_tokens_map.json", "tokenizer_file": null, "name_or_path": "facebook/m2m100_418M", "model_max_length": 1024, "additional_special_tokens": ["__af__", "__am__", "__ar__", "__ast__", "__az__", "__ba__", "__be__", "__bg__", "__bn__", "__br__", "__bs__", "__ca__", "__ceb__", "__cs__", "__cy__", "__da__", "__de__", "__el__", "__en__", "__es__", "__et__", "__fa__", "__ff__", "__fi__", "__fr__", "__fy__", "__ga__", "__gd__", "__gl__", "__gu__", "__ha__", "__he__", "__hi__", "__hr__", "__ht__", "__hu__", "__hy__", "__id__", "__ig__", "__ilo__", "__is__", "__it__", "__ja__", "__jv__", "__ka__", "__kk__", "__km__", "__kn__", "__ko__", "__lb__", "__lg__", "__ln__", "__lo__", "__lt__", "__lv__", "__mg__", "__mk__", "__ml__", "__mn__", "__mr__", "__ms__", "__my__", "__ne__", "__nl__", "__no__", "__ns__", "__oc__", "__or__", "__pa__", "__pl__", "__ps__", "__pt__", "__ro__", "__ru__", "__sd__", "__si__", "__sk__", "__sl__", "__so__", "__sq__", "__sr__", "__ss__", "__su__", "__sv__", "__sw__", "__ta__", "__th__", "__tl__", "__tn__", "__tr__", "__uk__", "__ur__", "__uz__", "__vi__", "__wo__", "__xh__", "__yi__", "__yo__", "__zh__", "__zu__"], "tokenizer_class": "M2M100Tokenizer"}
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 0.564786915779832,
4
+ "train_runtime": 39960.6259,
5
+ "train_samples": 663898,
6
+ "train_samples_per_second": 49.841,
7
+ "train_steps_per_second": 4.984
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,2413 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "global_step": 199170,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 4.98744790882161e-05,
13
+ "loss": 3.7619,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.02,
18
+ "learning_rate": 4.97489581764322e-05,
19
+ "loss": 2.8947,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.02,
24
+ "learning_rate": 4.962343726464829e-05,
25
+ "loss": 2.5182,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.03,
30
+ "learning_rate": 4.949791635286439e-05,
31
+ "loss": 2.3038,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.04,
36
+ "learning_rate": 4.937239544108049e-05,
37
+ "loss": 2.1453,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.05,
42
+ "learning_rate": 4.9246874529296584e-05,
43
+ "loss": 2.0297,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.05,
48
+ "learning_rate": 4.912135361751268e-05,
49
+ "loss": 1.9309,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.06,
54
+ "learning_rate": 4.899583270572878e-05,
55
+ "loss": 1.8563,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.07,
60
+ "learning_rate": 4.8870311793944875e-05,
61
+ "loss": 1.8049,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.08,
66
+ "learning_rate": 4.874479088216097e-05,
67
+ "loss": 1.7541,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.08,
72
+ "learning_rate": 4.861926997037707e-05,
73
+ "loss": 1.7037,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 0.09,
78
+ "learning_rate": 4.8493749058593166e-05,
79
+ "loss": 1.6645,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 0.1,
84
+ "learning_rate": 4.836822814680926e-05,
85
+ "loss": 1.621,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 0.11,
90
+ "learning_rate": 4.824270723502536e-05,
91
+ "loss": 1.6029,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 0.11,
96
+ "learning_rate": 4.8117186323241456e-05,
97
+ "loss": 1.5655,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 0.12,
102
+ "learning_rate": 4.799166541145755e-05,
103
+ "loss": 1.5463,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 0.13,
108
+ "learning_rate": 4.7866144499673646e-05,
109
+ "loss": 1.523,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 0.14,
114
+ "learning_rate": 4.774062358788974e-05,
115
+ "loss": 1.4924,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 0.14,
120
+ "learning_rate": 4.761510267610584e-05,
121
+ "loss": 1.4706,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 0.15,
126
+ "learning_rate": 4.7489581764321937e-05,
127
+ "loss": 1.4786,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 0.16,
132
+ "learning_rate": 4.736406085253803e-05,
133
+ "loss": 1.4326,
134
+ "step": 10500
135
+ },
136
+ {
137
+ "epoch": 0.17,
138
+ "learning_rate": 4.723853994075413e-05,
139
+ "loss": 1.4092,
140
+ "step": 11000
141
+ },
142
+ {
143
+ "epoch": 0.17,
144
+ "learning_rate": 4.711301902897023e-05,
145
+ "loss": 1.4245,
146
+ "step": 11500
147
+ },
148
+ {
149
+ "epoch": 0.18,
150
+ "learning_rate": 4.698749811718632e-05,
151
+ "loss": 1.4063,
152
+ "step": 12000
153
+ },
154
+ {
155
+ "epoch": 0.19,
156
+ "learning_rate": 4.6861977205402423e-05,
157
+ "loss": 1.3734,
158
+ "step": 12500
159
+ },
160
+ {
161
+ "epoch": 0.2,
162
+ "learning_rate": 4.673645629361852e-05,
163
+ "loss": 1.3613,
164
+ "step": 13000
165
+ },
166
+ {
167
+ "epoch": 0.2,
168
+ "learning_rate": 4.661093538183461e-05,
169
+ "loss": 1.3539,
170
+ "step": 13500
171
+ },
172
+ {
173
+ "epoch": 0.21,
174
+ "learning_rate": 4.6485414470050714e-05,
175
+ "loss": 1.3356,
176
+ "step": 14000
177
+ },
178
+ {
179
+ "epoch": 0.22,
180
+ "learning_rate": 4.635989355826681e-05,
181
+ "loss": 1.358,
182
+ "step": 14500
183
+ },
184
+ {
185
+ "epoch": 0.23,
186
+ "learning_rate": 4.6234372646482904e-05,
187
+ "loss": 1.311,
188
+ "step": 15000
189
+ },
190
+ {
191
+ "epoch": 0.23,
192
+ "learning_rate": 4.6108851734699005e-05,
193
+ "loss": 1.3026,
194
+ "step": 15500
195
+ },
196
+ {
197
+ "epoch": 0.24,
198
+ "learning_rate": 4.59833308229151e-05,
199
+ "loss": 1.308,
200
+ "step": 16000
201
+ },
202
+ {
203
+ "epoch": 0.25,
204
+ "learning_rate": 4.5857809911131194e-05,
205
+ "loss": 1.2758,
206
+ "step": 16500
207
+ },
208
+ {
209
+ "epoch": 0.26,
210
+ "learning_rate": 4.5732288999347296e-05,
211
+ "loss": 1.2894,
212
+ "step": 17000
213
+ },
214
+ {
215
+ "epoch": 0.26,
216
+ "learning_rate": 4.560676808756339e-05,
217
+ "loss": 1.2631,
218
+ "step": 17500
219
+ },
220
+ {
221
+ "epoch": 0.27,
222
+ "learning_rate": 4.5481247175779485e-05,
223
+ "loss": 1.2737,
224
+ "step": 18000
225
+ },
226
+ {
227
+ "epoch": 0.28,
228
+ "learning_rate": 4.535572626399559e-05,
229
+ "loss": 1.2568,
230
+ "step": 18500
231
+ },
232
+ {
233
+ "epoch": 0.29,
234
+ "learning_rate": 4.523020535221168e-05,
235
+ "loss": 1.2442,
236
+ "step": 19000
237
+ },
238
+ {
239
+ "epoch": 0.29,
240
+ "learning_rate": 4.510468444042778e-05,
241
+ "loss": 1.2455,
242
+ "step": 19500
243
+ },
244
+ {
245
+ "epoch": 0.3,
246
+ "learning_rate": 4.497916352864388e-05,
247
+ "loss": 1.244,
248
+ "step": 20000
249
+ },
250
+ {
251
+ "epoch": 0.31,
252
+ "learning_rate": 4.485364261685997e-05,
253
+ "loss": 1.23,
254
+ "step": 20500
255
+ },
256
+ {
257
+ "epoch": 0.32,
258
+ "learning_rate": 4.4728121705076074e-05,
259
+ "loss": 1.2089,
260
+ "step": 21000
261
+ },
262
+ {
263
+ "epoch": 0.32,
264
+ "learning_rate": 4.460260079329217e-05,
265
+ "loss": 1.2161,
266
+ "step": 21500
267
+ },
268
+ {
269
+ "epoch": 0.33,
270
+ "learning_rate": 4.447707988150826e-05,
271
+ "loss": 1.217,
272
+ "step": 22000
273
+ },
274
+ {
275
+ "epoch": 0.34,
276
+ "learning_rate": 4.435155896972436e-05,
277
+ "loss": 1.1856,
278
+ "step": 22500
279
+ },
280
+ {
281
+ "epoch": 0.35,
282
+ "learning_rate": 4.422603805794045e-05,
283
+ "loss": 1.1964,
284
+ "step": 23000
285
+ },
286
+ {
287
+ "epoch": 0.35,
288
+ "learning_rate": 4.410051714615655e-05,
289
+ "loss": 1.1793,
290
+ "step": 23500
291
+ },
292
+ {
293
+ "epoch": 0.36,
294
+ "learning_rate": 4.397499623437265e-05,
295
+ "loss": 1.1949,
296
+ "step": 24000
297
+ },
298
+ {
299
+ "epoch": 0.37,
300
+ "learning_rate": 4.384947532258874e-05,
301
+ "loss": 1.1823,
302
+ "step": 24500
303
+ },
304
+ {
305
+ "epoch": 0.38,
306
+ "learning_rate": 4.372395441080484e-05,
307
+ "loss": 1.1872,
308
+ "step": 25000
309
+ },
310
+ {
311
+ "epoch": 0.38,
312
+ "learning_rate": 4.359843349902094e-05,
313
+ "loss": 1.1712,
314
+ "step": 25500
315
+ },
316
+ {
317
+ "epoch": 0.39,
318
+ "learning_rate": 4.3472912587237034e-05,
319
+ "loss": 1.1792,
320
+ "step": 26000
321
+ },
322
+ {
323
+ "epoch": 0.4,
324
+ "learning_rate": 4.334739167545313e-05,
325
+ "loss": 1.1628,
326
+ "step": 26500
327
+ },
328
+ {
329
+ "epoch": 0.41,
330
+ "learning_rate": 4.322187076366923e-05,
331
+ "loss": 1.1464,
332
+ "step": 27000
333
+ },
334
+ {
335
+ "epoch": 0.41,
336
+ "learning_rate": 4.3096349851885325e-05,
337
+ "loss": 1.1668,
338
+ "step": 27500
339
+ },
340
+ {
341
+ "epoch": 0.42,
342
+ "learning_rate": 4.297082894010142e-05,
343
+ "loss": 1.1377,
344
+ "step": 28000
345
+ },
346
+ {
347
+ "epoch": 0.43,
348
+ "learning_rate": 4.284530802831752e-05,
349
+ "loss": 1.1426,
350
+ "step": 28500
351
+ },
352
+ {
353
+ "epoch": 0.44,
354
+ "learning_rate": 4.2719787116533615e-05,
355
+ "loss": 1.1347,
356
+ "step": 29000
357
+ },
358
+ {
359
+ "epoch": 0.44,
360
+ "learning_rate": 4.259426620474971e-05,
361
+ "loss": 1.1226,
362
+ "step": 29500
363
+ },
364
+ {
365
+ "epoch": 0.45,
366
+ "learning_rate": 4.246874529296581e-05,
367
+ "loss": 1.1094,
368
+ "step": 30000
369
+ },
370
+ {
371
+ "epoch": 0.46,
372
+ "learning_rate": 4.2343224381181906e-05,
373
+ "loss": 1.1204,
374
+ "step": 30500
375
+ },
376
+ {
377
+ "epoch": 0.47,
378
+ "learning_rate": 4.2217703469398e-05,
379
+ "loss": 1.1171,
380
+ "step": 31000
381
+ },
382
+ {
383
+ "epoch": 0.47,
384
+ "learning_rate": 4.20921825576141e-05,
385
+ "loss": 1.1354,
386
+ "step": 31500
387
+ },
388
+ {
389
+ "epoch": 0.48,
390
+ "learning_rate": 4.19666616458302e-05,
391
+ "loss": 1.0998,
392
+ "step": 32000
393
+ },
394
+ {
395
+ "epoch": 0.49,
396
+ "learning_rate": 4.18411407340463e-05,
397
+ "loss": 1.0948,
398
+ "step": 32500
399
+ },
400
+ {
401
+ "epoch": 0.5,
402
+ "learning_rate": 4.171561982226239e-05,
403
+ "loss": 1.0967,
404
+ "step": 33000
405
+ },
406
+ {
407
+ "epoch": 0.5,
408
+ "learning_rate": 4.159009891047849e-05,
409
+ "loss": 1.0943,
410
+ "step": 33500
411
+ },
412
+ {
413
+ "epoch": 0.51,
414
+ "learning_rate": 4.146457799869459e-05,
415
+ "loss": 1.1228,
416
+ "step": 34000
417
+ },
418
+ {
419
+ "epoch": 0.52,
420
+ "learning_rate": 4.1339057086910684e-05,
421
+ "loss": 1.0944,
422
+ "step": 34500
423
+ },
424
+ {
425
+ "epoch": 0.53,
426
+ "learning_rate": 4.121353617512678e-05,
427
+ "loss": 1.0928,
428
+ "step": 35000
429
+ },
430
+ {
431
+ "epoch": 0.53,
432
+ "learning_rate": 4.108801526334288e-05,
433
+ "loss": 1.1025,
434
+ "step": 35500
435
+ },
436
+ {
437
+ "epoch": 0.54,
438
+ "learning_rate": 4.0962494351558975e-05,
439
+ "loss": 1.0873,
440
+ "step": 36000
441
+ },
442
+ {
443
+ "epoch": 0.55,
444
+ "learning_rate": 4.083697343977507e-05,
445
+ "loss": 1.0977,
446
+ "step": 36500
447
+ },
448
+ {
449
+ "epoch": 0.56,
450
+ "learning_rate": 4.0711452527991164e-05,
451
+ "loss": 1.0785,
452
+ "step": 37000
453
+ },
454
+ {
455
+ "epoch": 0.56,
456
+ "learning_rate": 4.0585931616207266e-05,
457
+ "loss": 1.0754,
458
+ "step": 37500
459
+ },
460
+ {
461
+ "epoch": 0.57,
462
+ "learning_rate": 4.046041070442336e-05,
463
+ "loss": 1.0702,
464
+ "step": 38000
465
+ },
466
+ {
467
+ "epoch": 0.58,
468
+ "learning_rate": 4.0334889792639455e-05,
469
+ "loss": 1.0656,
470
+ "step": 38500
471
+ },
472
+ {
473
+ "epoch": 0.59,
474
+ "learning_rate": 4.020936888085555e-05,
475
+ "loss": 1.0817,
476
+ "step": 39000
477
+ },
478
+ {
479
+ "epoch": 0.59,
480
+ "learning_rate": 4.0083847969071644e-05,
481
+ "loss": 1.0647,
482
+ "step": 39500
483
+ },
484
+ {
485
+ "epoch": 0.6,
486
+ "learning_rate": 3.9958327057287746e-05,
487
+ "loss": 1.0403,
488
+ "step": 40000
489
+ },
490
+ {
491
+ "epoch": 0.61,
492
+ "learning_rate": 3.983280614550384e-05,
493
+ "loss": 1.052,
494
+ "step": 40500
495
+ },
496
+ {
497
+ "epoch": 0.62,
498
+ "learning_rate": 3.9707285233719935e-05,
499
+ "loss": 1.0472,
500
+ "step": 41000
501
+ },
502
+ {
503
+ "epoch": 0.63,
504
+ "learning_rate": 3.9581764321936036e-05,
505
+ "loss": 1.0509,
506
+ "step": 41500
507
+ },
508
+ {
509
+ "epoch": 0.63,
510
+ "learning_rate": 3.945624341015213e-05,
511
+ "loss": 1.0534,
512
+ "step": 42000
513
+ },
514
+ {
515
+ "epoch": 0.64,
516
+ "learning_rate": 3.9330722498368226e-05,
517
+ "loss": 1.0461,
518
+ "step": 42500
519
+ },
520
+ {
521
+ "epoch": 0.65,
522
+ "learning_rate": 3.920520158658433e-05,
523
+ "loss": 1.0376,
524
+ "step": 43000
525
+ },
526
+ {
527
+ "epoch": 0.66,
528
+ "learning_rate": 3.907968067480042e-05,
529
+ "loss": 1.0298,
530
+ "step": 43500
531
+ },
532
+ {
533
+ "epoch": 0.66,
534
+ "learning_rate": 3.895415976301652e-05,
535
+ "loss": 1.0333,
536
+ "step": 44000
537
+ },
538
+ {
539
+ "epoch": 0.67,
540
+ "learning_rate": 3.882863885123262e-05,
541
+ "loss": 1.0267,
542
+ "step": 44500
543
+ },
544
+ {
545
+ "epoch": 0.68,
546
+ "learning_rate": 3.870311793944871e-05,
547
+ "loss": 1.0263,
548
+ "step": 45000
549
+ },
550
+ {
551
+ "epoch": 0.69,
552
+ "learning_rate": 3.8577597027664814e-05,
553
+ "loss": 1.0276,
554
+ "step": 45500
555
+ },
556
+ {
557
+ "epoch": 0.69,
558
+ "learning_rate": 3.845207611588091e-05,
559
+ "loss": 1.0267,
560
+ "step": 46000
561
+ },
562
+ {
563
+ "epoch": 0.7,
564
+ "learning_rate": 3.8326555204097004e-05,
565
+ "loss": 1.0242,
566
+ "step": 46500
567
+ },
568
+ {
569
+ "epoch": 0.71,
570
+ "learning_rate": 3.8201034292313105e-05,
571
+ "loss": 1.0191,
572
+ "step": 47000
573
+ },
574
+ {
575
+ "epoch": 0.72,
576
+ "learning_rate": 3.80755133805292e-05,
577
+ "loss": 1.0134,
578
+ "step": 47500
579
+ },
580
+ {
581
+ "epoch": 0.72,
582
+ "learning_rate": 3.7949992468745294e-05,
583
+ "loss": 1.0165,
584
+ "step": 48000
585
+ },
586
+ {
587
+ "epoch": 0.73,
588
+ "learning_rate": 3.7824471556961396e-05,
589
+ "loss": 1.0201,
590
+ "step": 48500
591
+ },
592
+ {
593
+ "epoch": 0.74,
594
+ "learning_rate": 3.769895064517749e-05,
595
+ "loss": 1.0201,
596
+ "step": 49000
597
+ },
598
+ {
599
+ "epoch": 0.75,
600
+ "learning_rate": 3.7573429733393585e-05,
601
+ "loss": 1.02,
602
+ "step": 49500
603
+ },
604
+ {
605
+ "epoch": 0.75,
606
+ "learning_rate": 3.7447908821609687e-05,
607
+ "loss": 1.0144,
608
+ "step": 50000
609
+ },
610
+ {
611
+ "epoch": 0.76,
612
+ "learning_rate": 3.732238790982578e-05,
613
+ "loss": 0.9911,
614
+ "step": 50500
615
+ },
616
+ {
617
+ "epoch": 0.77,
618
+ "learning_rate": 3.7196866998041876e-05,
619
+ "loss": 0.9944,
620
+ "step": 51000
621
+ },
622
+ {
623
+ "epoch": 0.78,
624
+ "learning_rate": 3.707134608625798e-05,
625
+ "loss": 0.9963,
626
+ "step": 51500
627
+ },
628
+ {
629
+ "epoch": 0.78,
630
+ "learning_rate": 3.694582517447407e-05,
631
+ "loss": 0.9875,
632
+ "step": 52000
633
+ },
634
+ {
635
+ "epoch": 0.79,
636
+ "learning_rate": 3.682030426269017e-05,
637
+ "loss": 1.0059,
638
+ "step": 52500
639
+ },
640
+ {
641
+ "epoch": 0.8,
642
+ "learning_rate": 3.669478335090626e-05,
643
+ "loss": 0.9848,
644
+ "step": 53000
645
+ },
646
+ {
647
+ "epoch": 0.81,
648
+ "learning_rate": 3.6569262439122356e-05,
649
+ "loss": 0.9947,
650
+ "step": 53500
651
+ },
652
+ {
653
+ "epoch": 0.81,
654
+ "learning_rate": 3.644374152733845e-05,
655
+ "loss": 0.9865,
656
+ "step": 54000
657
+ },
658
+ {
659
+ "epoch": 0.82,
660
+ "learning_rate": 3.631822061555455e-05,
661
+ "loss": 0.9884,
662
+ "step": 54500
663
+ },
664
+ {
665
+ "epoch": 0.83,
666
+ "learning_rate": 3.619269970377065e-05,
667
+ "loss": 0.9781,
668
+ "step": 55000
669
+ },
670
+ {
671
+ "epoch": 0.84,
672
+ "learning_rate": 3.606717879198674e-05,
673
+ "loss": 0.9934,
674
+ "step": 55500
675
+ },
676
+ {
677
+ "epoch": 0.84,
678
+ "learning_rate": 3.594165788020284e-05,
679
+ "loss": 0.9937,
680
+ "step": 56000
681
+ },
682
+ {
683
+ "epoch": 0.85,
684
+ "learning_rate": 3.581613696841894e-05,
685
+ "loss": 0.9714,
686
+ "step": 56500
687
+ },
688
+ {
689
+ "epoch": 0.86,
690
+ "learning_rate": 3.569061605663503e-05,
691
+ "loss": 0.9809,
692
+ "step": 57000
693
+ },
694
+ {
695
+ "epoch": 0.87,
696
+ "learning_rate": 3.5565095144851134e-05,
697
+ "loss": 0.9709,
698
+ "step": 57500
699
+ },
700
+ {
701
+ "epoch": 0.87,
702
+ "learning_rate": 3.543957423306723e-05,
703
+ "loss": 0.9656,
704
+ "step": 58000
705
+ },
706
+ {
707
+ "epoch": 0.88,
708
+ "learning_rate": 3.531405332128333e-05,
709
+ "loss": 0.9632,
710
+ "step": 58500
711
+ },
712
+ {
713
+ "epoch": 0.89,
714
+ "learning_rate": 3.5188532409499425e-05,
715
+ "loss": 0.9767,
716
+ "step": 59000
717
+ },
718
+ {
719
+ "epoch": 0.9,
720
+ "learning_rate": 3.506301149771552e-05,
721
+ "loss": 0.9669,
722
+ "step": 59500
723
+ },
724
+ {
725
+ "epoch": 0.9,
726
+ "learning_rate": 3.493749058593162e-05,
727
+ "loss": 0.9692,
728
+ "step": 60000
729
+ },
730
+ {
731
+ "epoch": 0.91,
732
+ "learning_rate": 3.4811969674147715e-05,
733
+ "loss": 0.97,
734
+ "step": 60500
735
+ },
736
+ {
737
+ "epoch": 0.92,
738
+ "learning_rate": 3.468644876236381e-05,
739
+ "loss": 0.9514,
740
+ "step": 61000
741
+ },
742
+ {
743
+ "epoch": 0.93,
744
+ "learning_rate": 3.456092785057991e-05,
745
+ "loss": 0.9421,
746
+ "step": 61500
747
+ },
748
+ {
749
+ "epoch": 0.93,
750
+ "learning_rate": 3.4435406938796006e-05,
751
+ "loss": 0.9734,
752
+ "step": 62000
753
+ },
754
+ {
755
+ "epoch": 0.94,
756
+ "learning_rate": 3.43098860270121e-05,
757
+ "loss": 0.9562,
758
+ "step": 62500
759
+ },
760
+ {
761
+ "epoch": 0.95,
762
+ "learning_rate": 3.41843651152282e-05,
763
+ "loss": 0.9625,
764
+ "step": 63000
765
+ },
766
+ {
767
+ "epoch": 0.96,
768
+ "learning_rate": 3.40588442034443e-05,
769
+ "loss": 0.9542,
770
+ "step": 63500
771
+ },
772
+ {
773
+ "epoch": 0.96,
774
+ "learning_rate": 3.393332329166039e-05,
775
+ "loss": 0.9497,
776
+ "step": 64000
777
+ },
778
+ {
779
+ "epoch": 0.97,
780
+ "learning_rate": 3.380780237987649e-05,
781
+ "loss": 0.9431,
782
+ "step": 64500
783
+ },
784
+ {
785
+ "epoch": 0.98,
786
+ "learning_rate": 3.368228146809259e-05,
787
+ "loss": 0.9456,
788
+ "step": 65000
789
+ },
790
+ {
791
+ "epoch": 0.99,
792
+ "learning_rate": 3.355676055630868e-05,
793
+ "loss": 0.9395,
794
+ "step": 65500
795
+ },
796
+ {
797
+ "epoch": 0.99,
798
+ "learning_rate": 3.3431239644524784e-05,
799
+ "loss": 0.9393,
800
+ "step": 66000
801
+ },
802
+ {
803
+ "epoch": 1.0,
804
+ "learning_rate": 3.330571873274088e-05,
805
+ "loss": 0.9243,
806
+ "step": 66500
807
+ },
808
+ {
809
+ "epoch": 1.01,
810
+ "learning_rate": 3.318019782095697e-05,
811
+ "loss": 0.8449,
812
+ "step": 67000
813
+ },
814
+ {
815
+ "epoch": 1.02,
816
+ "learning_rate": 3.305467690917307e-05,
817
+ "loss": 0.8542,
818
+ "step": 67500
819
+ },
820
+ {
821
+ "epoch": 1.02,
822
+ "learning_rate": 3.292915599738917e-05,
823
+ "loss": 0.8496,
824
+ "step": 68000
825
+ },
826
+ {
827
+ "epoch": 1.03,
828
+ "learning_rate": 3.2803635085605264e-05,
829
+ "loss": 0.8452,
830
+ "step": 68500
831
+ },
832
+ {
833
+ "epoch": 1.04,
834
+ "learning_rate": 3.267811417382136e-05,
835
+ "loss": 0.8361,
836
+ "step": 69000
837
+ },
838
+ {
839
+ "epoch": 1.05,
840
+ "learning_rate": 3.2552593262037453e-05,
841
+ "loss": 0.8441,
842
+ "step": 69500
843
+ },
844
+ {
845
+ "epoch": 1.05,
846
+ "learning_rate": 3.242707235025355e-05,
847
+ "loss": 0.8536,
848
+ "step": 70000
849
+ },
850
+ {
851
+ "epoch": 1.06,
852
+ "learning_rate": 3.230155143846965e-05,
853
+ "loss": 0.8338,
854
+ "step": 70500
855
+ },
856
+ {
857
+ "epoch": 1.07,
858
+ "learning_rate": 3.2176030526685744e-05,
859
+ "loss": 0.845,
860
+ "step": 71000
861
+ },
862
+ {
863
+ "epoch": 1.08,
864
+ "learning_rate": 3.2050509614901846e-05,
865
+ "loss": 0.8544,
866
+ "step": 71500
867
+ },
868
+ {
869
+ "epoch": 1.08,
870
+ "learning_rate": 3.192498870311794e-05,
871
+ "loss": 0.8487,
872
+ "step": 72000
873
+ },
874
+ {
875
+ "epoch": 1.09,
876
+ "learning_rate": 3.1799467791334035e-05,
877
+ "loss": 0.8273,
878
+ "step": 72500
879
+ },
880
+ {
881
+ "epoch": 1.1,
882
+ "learning_rate": 3.1673946879550136e-05,
883
+ "loss": 0.8581,
884
+ "step": 73000
885
+ },
886
+ {
887
+ "epoch": 1.11,
888
+ "learning_rate": 3.154842596776623e-05,
889
+ "loss": 0.8493,
890
+ "step": 73500
891
+ },
892
+ {
893
+ "epoch": 1.11,
894
+ "learning_rate": 3.1422905055982326e-05,
895
+ "loss": 0.8321,
896
+ "step": 74000
897
+ },
898
+ {
899
+ "epoch": 1.12,
900
+ "learning_rate": 3.129738414419843e-05,
901
+ "loss": 0.8404,
902
+ "step": 74500
903
+ },
904
+ {
905
+ "epoch": 1.13,
906
+ "learning_rate": 3.117186323241452e-05,
907
+ "loss": 0.8207,
908
+ "step": 75000
909
+ },
910
+ {
911
+ "epoch": 1.14,
912
+ "learning_rate": 3.104634232063062e-05,
913
+ "loss": 0.8439,
914
+ "step": 75500
915
+ },
916
+ {
917
+ "epoch": 1.14,
918
+ "learning_rate": 3.092082140884672e-05,
919
+ "loss": 0.8462,
920
+ "step": 76000
921
+ },
922
+ {
923
+ "epoch": 1.15,
924
+ "learning_rate": 3.079530049706281e-05,
925
+ "loss": 0.8441,
926
+ "step": 76500
927
+ },
928
+ {
929
+ "epoch": 1.16,
930
+ "learning_rate": 3.066977958527891e-05,
931
+ "loss": 0.8394,
932
+ "step": 77000
933
+ },
934
+ {
935
+ "epoch": 1.17,
936
+ "learning_rate": 3.054425867349501e-05,
937
+ "loss": 0.8441,
938
+ "step": 77500
939
+ },
940
+ {
941
+ "epoch": 1.17,
942
+ "learning_rate": 3.0418737761711104e-05,
943
+ "loss": 0.8462,
944
+ "step": 78000
945
+ },
946
+ {
947
+ "epoch": 1.18,
948
+ "learning_rate": 3.0293216849927198e-05,
949
+ "loss": 0.8392,
950
+ "step": 78500
951
+ },
952
+ {
953
+ "epoch": 1.19,
954
+ "learning_rate": 3.0167695938143296e-05,
955
+ "loss": 0.835,
956
+ "step": 79000
957
+ },
958
+ {
959
+ "epoch": 1.2,
960
+ "learning_rate": 3.004217502635939e-05,
961
+ "loss": 0.8544,
962
+ "step": 79500
963
+ },
964
+ {
965
+ "epoch": 1.21,
966
+ "learning_rate": 2.991665411457549e-05,
967
+ "loss": 0.8275,
968
+ "step": 80000
969
+ },
970
+ {
971
+ "epoch": 1.21,
972
+ "learning_rate": 2.9791133202791587e-05,
973
+ "loss": 0.8288,
974
+ "step": 80500
975
+ },
976
+ {
977
+ "epoch": 1.22,
978
+ "learning_rate": 2.9665612291007682e-05,
979
+ "loss": 0.838,
980
+ "step": 81000
981
+ },
982
+ {
983
+ "epoch": 1.23,
984
+ "learning_rate": 2.9540091379223776e-05,
985
+ "loss": 0.8369,
986
+ "step": 81500
987
+ },
988
+ {
989
+ "epoch": 1.24,
990
+ "learning_rate": 2.9414570467439878e-05,
991
+ "loss": 0.8309,
992
+ "step": 82000
993
+ },
994
+ {
995
+ "epoch": 1.24,
996
+ "learning_rate": 2.9289049555655973e-05,
997
+ "loss": 0.8313,
998
+ "step": 82500
999
+ },
1000
+ {
1001
+ "epoch": 1.25,
1002
+ "learning_rate": 2.9163528643872067e-05,
1003
+ "loss": 0.845,
1004
+ "step": 83000
1005
+ },
1006
+ {
1007
+ "epoch": 1.26,
1008
+ "learning_rate": 2.903800773208817e-05,
1009
+ "loss": 0.8348,
1010
+ "step": 83500
1011
+ },
1012
+ {
1013
+ "epoch": 1.27,
1014
+ "learning_rate": 2.8912486820304263e-05,
1015
+ "loss": 0.8231,
1016
+ "step": 84000
1017
+ },
1018
+ {
1019
+ "epoch": 1.27,
1020
+ "learning_rate": 2.8786965908520365e-05,
1021
+ "loss": 0.8286,
1022
+ "step": 84500
1023
+ },
1024
+ {
1025
+ "epoch": 1.28,
1026
+ "learning_rate": 2.866144499673646e-05,
1027
+ "loss": 0.8203,
1028
+ "step": 85000
1029
+ },
1030
+ {
1031
+ "epoch": 1.29,
1032
+ "learning_rate": 2.8535924084952554e-05,
1033
+ "loss": 0.8264,
1034
+ "step": 85500
1035
+ },
1036
+ {
1037
+ "epoch": 1.3,
1038
+ "learning_rate": 2.8410403173168652e-05,
1039
+ "loss": 0.8291,
1040
+ "step": 86000
1041
+ },
1042
+ {
1043
+ "epoch": 1.3,
1044
+ "learning_rate": 2.828488226138475e-05,
1045
+ "loss": 0.8222,
1046
+ "step": 86500
1047
+ },
1048
+ {
1049
+ "epoch": 1.31,
1050
+ "learning_rate": 2.8159361349600845e-05,
1051
+ "loss": 0.8357,
1052
+ "step": 87000
1053
+ },
1054
+ {
1055
+ "epoch": 1.32,
1056
+ "learning_rate": 2.8033840437816943e-05,
1057
+ "loss": 0.8327,
1058
+ "step": 87500
1059
+ },
1060
+ {
1061
+ "epoch": 1.33,
1062
+ "learning_rate": 2.7908319526033038e-05,
1063
+ "loss": 0.834,
1064
+ "step": 88000
1065
+ },
1066
+ {
1067
+ "epoch": 1.33,
1068
+ "learning_rate": 2.7782798614249132e-05,
1069
+ "loss": 0.8187,
1070
+ "step": 88500
1071
+ },
1072
+ {
1073
+ "epoch": 1.34,
1074
+ "learning_rate": 2.7657277702465234e-05,
1075
+ "loss": 0.8252,
1076
+ "step": 89000
1077
+ },
1078
+ {
1079
+ "epoch": 1.35,
1080
+ "learning_rate": 2.753175679068133e-05,
1081
+ "loss": 0.8313,
1082
+ "step": 89500
1083
+ },
1084
+ {
1085
+ "epoch": 1.36,
1086
+ "learning_rate": 2.7406235878897423e-05,
1087
+ "loss": 0.8255,
1088
+ "step": 90000
1089
+ },
1090
+ {
1091
+ "epoch": 1.36,
1092
+ "learning_rate": 2.7280714967113525e-05,
1093
+ "loss": 0.827,
1094
+ "step": 90500
1095
+ },
1096
+ {
1097
+ "epoch": 1.37,
1098
+ "learning_rate": 2.715519405532962e-05,
1099
+ "loss": 0.8251,
1100
+ "step": 91000
1101
+ },
1102
+ {
1103
+ "epoch": 1.38,
1104
+ "learning_rate": 2.7029673143545714e-05,
1105
+ "loss": 0.8173,
1106
+ "step": 91500
1107
+ },
1108
+ {
1109
+ "epoch": 1.39,
1110
+ "learning_rate": 2.6904152231761815e-05,
1111
+ "loss": 0.82,
1112
+ "step": 92000
1113
+ },
1114
+ {
1115
+ "epoch": 1.39,
1116
+ "learning_rate": 2.677863131997791e-05,
1117
+ "loss": 0.814,
1118
+ "step": 92500
1119
+ },
1120
+ {
1121
+ "epoch": 1.4,
1122
+ "learning_rate": 2.6653110408194005e-05,
1123
+ "loss": 0.8218,
1124
+ "step": 93000
1125
+ },
1126
+ {
1127
+ "epoch": 1.41,
1128
+ "learning_rate": 2.6527589496410106e-05,
1129
+ "loss": 0.8118,
1130
+ "step": 93500
1131
+ },
1132
+ {
1133
+ "epoch": 1.42,
1134
+ "learning_rate": 2.64020685846262e-05,
1135
+ "loss": 0.8189,
1136
+ "step": 94000
1137
+ },
1138
+ {
1139
+ "epoch": 1.42,
1140
+ "learning_rate": 2.6276547672842296e-05,
1141
+ "loss": 0.8215,
1142
+ "step": 94500
1143
+ },
1144
+ {
1145
+ "epoch": 1.43,
1146
+ "learning_rate": 2.6151026761058394e-05,
1147
+ "loss": 0.822,
1148
+ "step": 95000
1149
+ },
1150
+ {
1151
+ "epoch": 1.44,
1152
+ "learning_rate": 2.6025505849274488e-05,
1153
+ "loss": 0.8149,
1154
+ "step": 95500
1155
+ },
1156
+ {
1157
+ "epoch": 1.45,
1158
+ "learning_rate": 2.5899984937490583e-05,
1159
+ "loss": 0.8271,
1160
+ "step": 96000
1161
+ },
1162
+ {
1163
+ "epoch": 1.45,
1164
+ "learning_rate": 2.5774464025706684e-05,
1165
+ "loss": 0.823,
1166
+ "step": 96500
1167
+ },
1168
+ {
1169
+ "epoch": 1.46,
1170
+ "learning_rate": 2.564894311392278e-05,
1171
+ "loss": 0.8167,
1172
+ "step": 97000
1173
+ },
1174
+ {
1175
+ "epoch": 1.47,
1176
+ "learning_rate": 2.552342220213888e-05,
1177
+ "loss": 0.8127,
1178
+ "step": 97500
1179
+ },
1180
+ {
1181
+ "epoch": 1.48,
1182
+ "learning_rate": 2.5397901290354975e-05,
1183
+ "loss": 0.7968,
1184
+ "step": 98000
1185
+ },
1186
+ {
1187
+ "epoch": 1.48,
1188
+ "learning_rate": 2.527238037857107e-05,
1189
+ "loss": 0.8144,
1190
+ "step": 98500
1191
+ },
1192
+ {
1193
+ "epoch": 1.49,
1194
+ "learning_rate": 2.514685946678717e-05,
1195
+ "loss": 0.8111,
1196
+ "step": 99000
1197
+ },
1198
+ {
1199
+ "epoch": 1.5,
1200
+ "learning_rate": 2.5021338555003266e-05,
1201
+ "loss": 0.8125,
1202
+ "step": 99500
1203
+ },
1204
+ {
1205
+ "epoch": 1.51,
1206
+ "learning_rate": 2.489581764321936e-05,
1207
+ "loss": 0.8158,
1208
+ "step": 100000
1209
+ },
1210
+ {
1211
+ "epoch": 1.51,
1212
+ "learning_rate": 2.477029673143546e-05,
1213
+ "loss": 0.8132,
1214
+ "step": 100500
1215
+ },
1216
+ {
1217
+ "epoch": 1.52,
1218
+ "learning_rate": 2.4644775819651557e-05,
1219
+ "loss": 0.8119,
1220
+ "step": 101000
1221
+ },
1222
+ {
1223
+ "epoch": 1.53,
1224
+ "learning_rate": 2.451925490786765e-05,
1225
+ "loss": 0.8121,
1226
+ "step": 101500
1227
+ },
1228
+ {
1229
+ "epoch": 1.54,
1230
+ "learning_rate": 2.439373399608375e-05,
1231
+ "loss": 0.7989,
1232
+ "step": 102000
1233
+ },
1234
+ {
1235
+ "epoch": 1.54,
1236
+ "learning_rate": 2.4268213084299844e-05,
1237
+ "loss": 0.8099,
1238
+ "step": 102500
1239
+ },
1240
+ {
1241
+ "epoch": 1.55,
1242
+ "learning_rate": 2.4142692172515942e-05,
1243
+ "loss": 0.8157,
1244
+ "step": 103000
1245
+ },
1246
+ {
1247
+ "epoch": 1.56,
1248
+ "learning_rate": 2.4017171260732037e-05,
1249
+ "loss": 0.807,
1250
+ "step": 103500
1251
+ },
1252
+ {
1253
+ "epoch": 1.57,
1254
+ "learning_rate": 2.3891650348948135e-05,
1255
+ "loss": 0.8021,
1256
+ "step": 104000
1257
+ },
1258
+ {
1259
+ "epoch": 1.57,
1260
+ "learning_rate": 2.3766129437164233e-05,
1261
+ "loss": 0.7899,
1262
+ "step": 104500
1263
+ },
1264
+ {
1265
+ "epoch": 1.58,
1266
+ "learning_rate": 2.3640608525380328e-05,
1267
+ "loss": 0.8051,
1268
+ "step": 105000
1269
+ },
1270
+ {
1271
+ "epoch": 1.59,
1272
+ "learning_rate": 2.3515087613596426e-05,
1273
+ "loss": 0.8025,
1274
+ "step": 105500
1275
+ },
1276
+ {
1277
+ "epoch": 1.6,
1278
+ "learning_rate": 2.3389566701812524e-05,
1279
+ "loss": 0.7956,
1280
+ "step": 106000
1281
+ },
1282
+ {
1283
+ "epoch": 1.6,
1284
+ "learning_rate": 2.326404579002862e-05,
1285
+ "loss": 0.8237,
1286
+ "step": 106500
1287
+ },
1288
+ {
1289
+ "epoch": 1.61,
1290
+ "learning_rate": 2.3138524878244717e-05,
1291
+ "loss": 0.8056,
1292
+ "step": 107000
1293
+ },
1294
+ {
1295
+ "epoch": 1.62,
1296
+ "learning_rate": 2.3013003966460815e-05,
1297
+ "loss": 0.7946,
1298
+ "step": 107500
1299
+ },
1300
+ {
1301
+ "epoch": 1.63,
1302
+ "learning_rate": 2.2887483054676913e-05,
1303
+ "loss": 0.7888,
1304
+ "step": 108000
1305
+ },
1306
+ {
1307
+ "epoch": 1.63,
1308
+ "learning_rate": 2.2761962142893007e-05,
1309
+ "loss": 0.7957,
1310
+ "step": 108500
1311
+ },
1312
+ {
1313
+ "epoch": 1.64,
1314
+ "learning_rate": 2.2636441231109105e-05,
1315
+ "loss": 0.7968,
1316
+ "step": 109000
1317
+ },
1318
+ {
1319
+ "epoch": 1.65,
1320
+ "learning_rate": 2.25109203193252e-05,
1321
+ "loss": 0.7996,
1322
+ "step": 109500
1323
+ },
1324
+ {
1325
+ "epoch": 1.66,
1326
+ "learning_rate": 2.2385399407541298e-05,
1327
+ "loss": 0.7899,
1328
+ "step": 110000
1329
+ },
1330
+ {
1331
+ "epoch": 1.66,
1332
+ "learning_rate": 2.2259878495757393e-05,
1333
+ "loss": 0.8012,
1334
+ "step": 110500
1335
+ },
1336
+ {
1337
+ "epoch": 1.67,
1338
+ "learning_rate": 2.213435758397349e-05,
1339
+ "loss": 0.7944,
1340
+ "step": 111000
1341
+ },
1342
+ {
1343
+ "epoch": 1.68,
1344
+ "learning_rate": 2.2008836672189586e-05,
1345
+ "loss": 0.7916,
1346
+ "step": 111500
1347
+ },
1348
+ {
1349
+ "epoch": 1.69,
1350
+ "learning_rate": 2.1883315760405684e-05,
1351
+ "loss": 0.7932,
1352
+ "step": 112000
1353
+ },
1354
+ {
1355
+ "epoch": 1.69,
1356
+ "learning_rate": 2.1757794848621782e-05,
1357
+ "loss": 0.7904,
1358
+ "step": 112500
1359
+ },
1360
+ {
1361
+ "epoch": 1.7,
1362
+ "learning_rate": 2.1632273936837876e-05,
1363
+ "loss": 0.7935,
1364
+ "step": 113000
1365
+ },
1366
+ {
1367
+ "epoch": 1.71,
1368
+ "learning_rate": 2.1506753025053974e-05,
1369
+ "loss": 0.7881,
1370
+ "step": 113500
1371
+ },
1372
+ {
1373
+ "epoch": 1.72,
1374
+ "learning_rate": 2.1381232113270073e-05,
1375
+ "loss": 0.7754,
1376
+ "step": 114000
1377
+ },
1378
+ {
1379
+ "epoch": 1.72,
1380
+ "learning_rate": 2.125571120148617e-05,
1381
+ "loss": 0.7974,
1382
+ "step": 114500
1383
+ },
1384
+ {
1385
+ "epoch": 1.73,
1386
+ "learning_rate": 2.1130190289702265e-05,
1387
+ "loss": 0.7857,
1388
+ "step": 115000
1389
+ },
1390
+ {
1391
+ "epoch": 1.74,
1392
+ "learning_rate": 2.1004669377918363e-05,
1393
+ "loss": 0.7714,
1394
+ "step": 115500
1395
+ },
1396
+ {
1397
+ "epoch": 1.75,
1398
+ "learning_rate": 2.087914846613446e-05,
1399
+ "loss": 0.7925,
1400
+ "step": 116000
1401
+ },
1402
+ {
1403
+ "epoch": 1.75,
1404
+ "learning_rate": 2.0753627554350556e-05,
1405
+ "loss": 0.7864,
1406
+ "step": 116500
1407
+ },
1408
+ {
1409
+ "epoch": 1.76,
1410
+ "learning_rate": 2.0628106642566654e-05,
1411
+ "loss": 0.7866,
1412
+ "step": 117000
1413
+ },
1414
+ {
1415
+ "epoch": 1.77,
1416
+ "learning_rate": 2.050258573078275e-05,
1417
+ "loss": 0.789,
1418
+ "step": 117500
1419
+ },
1420
+ {
1421
+ "epoch": 1.78,
1422
+ "learning_rate": 2.0377064818998843e-05,
1423
+ "loss": 0.7667,
1424
+ "step": 118000
1425
+ },
1426
+ {
1427
+ "epoch": 1.78,
1428
+ "learning_rate": 2.025154390721494e-05,
1429
+ "loss": 0.7824,
1430
+ "step": 118500
1431
+ },
1432
+ {
1433
+ "epoch": 1.79,
1434
+ "learning_rate": 2.012602299543104e-05,
1435
+ "loss": 0.7811,
1436
+ "step": 119000
1437
+ },
1438
+ {
1439
+ "epoch": 1.8,
1440
+ "learning_rate": 2.0000502083647138e-05,
1441
+ "loss": 0.7861,
1442
+ "step": 119500
1443
+ },
1444
+ {
1445
+ "epoch": 1.81,
1446
+ "learning_rate": 1.9874981171863232e-05,
1447
+ "loss": 0.779,
1448
+ "step": 120000
1449
+ },
1450
+ {
1451
+ "epoch": 1.82,
1452
+ "learning_rate": 1.974946026007933e-05,
1453
+ "loss": 0.7766,
1454
+ "step": 120500
1455
+ },
1456
+ {
1457
+ "epoch": 1.82,
1458
+ "learning_rate": 1.962393934829543e-05,
1459
+ "loss": 0.772,
1460
+ "step": 121000
1461
+ },
1462
+ {
1463
+ "epoch": 1.83,
1464
+ "learning_rate": 1.9498418436511523e-05,
1465
+ "loss": 0.7826,
1466
+ "step": 121500
1467
+ },
1468
+ {
1469
+ "epoch": 1.84,
1470
+ "learning_rate": 1.937289752472762e-05,
1471
+ "loss": 0.7771,
1472
+ "step": 122000
1473
+ },
1474
+ {
1475
+ "epoch": 1.85,
1476
+ "learning_rate": 1.924737661294372e-05,
1477
+ "loss": 0.7636,
1478
+ "step": 122500
1479
+ },
1480
+ {
1481
+ "epoch": 1.85,
1482
+ "learning_rate": 1.9121855701159814e-05,
1483
+ "loss": 0.7749,
1484
+ "step": 123000
1485
+ },
1486
+ {
1487
+ "epoch": 1.86,
1488
+ "learning_rate": 1.8996334789375912e-05,
1489
+ "loss": 0.7696,
1490
+ "step": 123500
1491
+ },
1492
+ {
1493
+ "epoch": 1.87,
1494
+ "learning_rate": 1.887081387759201e-05,
1495
+ "loss": 0.7742,
1496
+ "step": 124000
1497
+ },
1498
+ {
1499
+ "epoch": 1.88,
1500
+ "learning_rate": 1.8745292965808105e-05,
1501
+ "loss": 0.7568,
1502
+ "step": 124500
1503
+ },
1504
+ {
1505
+ "epoch": 1.88,
1506
+ "learning_rate": 1.86197720540242e-05,
1507
+ "loss": 0.7782,
1508
+ "step": 125000
1509
+ },
1510
+ {
1511
+ "epoch": 1.89,
1512
+ "learning_rate": 1.8494251142240297e-05,
1513
+ "loss": 0.7762,
1514
+ "step": 125500
1515
+ },
1516
+ {
1517
+ "epoch": 1.9,
1518
+ "learning_rate": 1.8368730230456396e-05,
1519
+ "loss": 0.7607,
1520
+ "step": 126000
1521
+ },
1522
+ {
1523
+ "epoch": 1.91,
1524
+ "learning_rate": 1.824320931867249e-05,
1525
+ "loss": 0.764,
1526
+ "step": 126500
1527
+ },
1528
+ {
1529
+ "epoch": 1.91,
1530
+ "learning_rate": 1.8117688406888588e-05,
1531
+ "loss": 0.7698,
1532
+ "step": 127000
1533
+ },
1534
+ {
1535
+ "epoch": 1.92,
1536
+ "learning_rate": 1.7992167495104686e-05,
1537
+ "loss": 0.768,
1538
+ "step": 127500
1539
+ },
1540
+ {
1541
+ "epoch": 1.93,
1542
+ "learning_rate": 1.786664658332078e-05,
1543
+ "loss": 0.7729,
1544
+ "step": 128000
1545
+ },
1546
+ {
1547
+ "epoch": 1.94,
1548
+ "learning_rate": 1.774112567153688e-05,
1549
+ "loss": 0.7641,
1550
+ "step": 128500
1551
+ },
1552
+ {
1553
+ "epoch": 1.94,
1554
+ "learning_rate": 1.7615604759752977e-05,
1555
+ "loss": 0.7623,
1556
+ "step": 129000
1557
+ },
1558
+ {
1559
+ "epoch": 1.95,
1560
+ "learning_rate": 1.7490083847969072e-05,
1561
+ "loss": 0.7729,
1562
+ "step": 129500
1563
+ },
1564
+ {
1565
+ "epoch": 1.96,
1566
+ "learning_rate": 1.736456293618517e-05,
1567
+ "loss": 0.7548,
1568
+ "step": 130000
1569
+ },
1570
+ {
1571
+ "epoch": 1.97,
1572
+ "learning_rate": 1.7239042024401268e-05,
1573
+ "loss": 0.7738,
1574
+ "step": 130500
1575
+ },
1576
+ {
1577
+ "epoch": 1.97,
1578
+ "learning_rate": 1.7113521112617363e-05,
1579
+ "loss": 0.7659,
1580
+ "step": 131000
1581
+ },
1582
+ {
1583
+ "epoch": 1.98,
1584
+ "learning_rate": 1.698800020083346e-05,
1585
+ "loss": 0.7525,
1586
+ "step": 131500
1587
+ },
1588
+ {
1589
+ "epoch": 1.99,
1590
+ "learning_rate": 1.6862479289049555e-05,
1591
+ "loss": 0.7809,
1592
+ "step": 132000
1593
+ },
1594
+ {
1595
+ "epoch": 2.0,
1596
+ "learning_rate": 1.6736958377265653e-05,
1597
+ "loss": 0.777,
1598
+ "step": 132500
1599
+ },
1600
+ {
1601
+ "epoch": 2.0,
1602
+ "learning_rate": 1.6611437465481748e-05,
1603
+ "loss": 0.725,
1604
+ "step": 133000
1605
+ },
1606
+ {
1607
+ "epoch": 2.01,
1608
+ "learning_rate": 1.6485916553697846e-05,
1609
+ "loss": 0.657,
1610
+ "step": 133500
1611
+ },
1612
+ {
1613
+ "epoch": 2.02,
1614
+ "learning_rate": 1.6360395641913944e-05,
1615
+ "loss": 0.6593,
1616
+ "step": 134000
1617
+ },
1618
+ {
1619
+ "epoch": 2.03,
1620
+ "learning_rate": 1.623487473013004e-05,
1621
+ "loss": 0.6657,
1622
+ "step": 134500
1623
+ },
1624
+ {
1625
+ "epoch": 2.03,
1626
+ "learning_rate": 1.6109353818346137e-05,
1627
+ "loss": 0.6536,
1628
+ "step": 135000
1629
+ },
1630
+ {
1631
+ "epoch": 2.04,
1632
+ "learning_rate": 1.5983832906562235e-05,
1633
+ "loss": 0.6721,
1634
+ "step": 135500
1635
+ },
1636
+ {
1637
+ "epoch": 2.05,
1638
+ "learning_rate": 1.585831199477833e-05,
1639
+ "loss": 0.6559,
1640
+ "step": 136000
1641
+ },
1642
+ {
1643
+ "epoch": 2.06,
1644
+ "learning_rate": 1.5732791082994428e-05,
1645
+ "loss": 0.6645,
1646
+ "step": 136500
1647
+ },
1648
+ {
1649
+ "epoch": 2.06,
1650
+ "learning_rate": 1.5607270171210526e-05,
1651
+ "loss": 0.6506,
1652
+ "step": 137000
1653
+ },
1654
+ {
1655
+ "epoch": 2.07,
1656
+ "learning_rate": 1.548174925942662e-05,
1657
+ "loss": 0.673,
1658
+ "step": 137500
1659
+ },
1660
+ {
1661
+ "epoch": 2.08,
1662
+ "learning_rate": 1.535622834764272e-05,
1663
+ "loss": 0.6579,
1664
+ "step": 138000
1665
+ },
1666
+ {
1667
+ "epoch": 2.09,
1668
+ "learning_rate": 1.5230707435858815e-05,
1669
+ "loss": 0.6589,
1670
+ "step": 138500
1671
+ },
1672
+ {
1673
+ "epoch": 2.09,
1674
+ "learning_rate": 1.5105186524074913e-05,
1675
+ "loss": 0.6698,
1676
+ "step": 139000
1677
+ },
1678
+ {
1679
+ "epoch": 2.1,
1680
+ "learning_rate": 1.4979665612291008e-05,
1681
+ "loss": 0.6613,
1682
+ "step": 139500
1683
+ },
1684
+ {
1685
+ "epoch": 2.11,
1686
+ "learning_rate": 1.4854144700507106e-05,
1687
+ "loss": 0.6528,
1688
+ "step": 140000
1689
+ },
1690
+ {
1691
+ "epoch": 2.12,
1692
+ "learning_rate": 1.4728623788723204e-05,
1693
+ "loss": 0.6667,
1694
+ "step": 140500
1695
+ },
1696
+ {
1697
+ "epoch": 2.12,
1698
+ "learning_rate": 1.4603102876939298e-05,
1699
+ "loss": 0.6694,
1700
+ "step": 141000
1701
+ },
1702
+ {
1703
+ "epoch": 2.13,
1704
+ "learning_rate": 1.4477581965155396e-05,
1705
+ "loss": 0.6735,
1706
+ "step": 141500
1707
+ },
1708
+ {
1709
+ "epoch": 2.14,
1710
+ "learning_rate": 1.4352061053371493e-05,
1711
+ "loss": 0.6559,
1712
+ "step": 142000
1713
+ },
1714
+ {
1715
+ "epoch": 2.15,
1716
+ "learning_rate": 1.4226540141587587e-05,
1717
+ "loss": 0.6551,
1718
+ "step": 142500
1719
+ },
1720
+ {
1721
+ "epoch": 2.15,
1722
+ "learning_rate": 1.4101019229803686e-05,
1723
+ "loss": 0.6628,
1724
+ "step": 143000
1725
+ },
1726
+ {
1727
+ "epoch": 2.16,
1728
+ "learning_rate": 1.3975498318019784e-05,
1729
+ "loss": 0.6638,
1730
+ "step": 143500
1731
+ },
1732
+ {
1733
+ "epoch": 2.17,
1734
+ "learning_rate": 1.3849977406235878e-05,
1735
+ "loss": 0.6602,
1736
+ "step": 144000
1737
+ },
1738
+ {
1739
+ "epoch": 2.18,
1740
+ "learning_rate": 1.3724456494451976e-05,
1741
+ "loss": 0.6618,
1742
+ "step": 144500
1743
+ },
1744
+ {
1745
+ "epoch": 2.18,
1746
+ "learning_rate": 1.3598935582668074e-05,
1747
+ "loss": 0.6663,
1748
+ "step": 145000
1749
+ },
1750
+ {
1751
+ "epoch": 2.19,
1752
+ "learning_rate": 1.347341467088417e-05,
1753
+ "loss": 0.6533,
1754
+ "step": 145500
1755
+ },
1756
+ {
1757
+ "epoch": 2.2,
1758
+ "learning_rate": 1.3347893759100265e-05,
1759
+ "loss": 0.6451,
1760
+ "step": 146000
1761
+ },
1762
+ {
1763
+ "epoch": 2.21,
1764
+ "learning_rate": 1.3222372847316364e-05,
1765
+ "loss": 0.6701,
1766
+ "step": 146500
1767
+ },
1768
+ {
1769
+ "epoch": 2.21,
1770
+ "learning_rate": 1.3096851935532462e-05,
1771
+ "loss": 0.6705,
1772
+ "step": 147000
1773
+ },
1774
+ {
1775
+ "epoch": 2.22,
1776
+ "learning_rate": 1.2971331023748556e-05,
1777
+ "loss": 0.6605,
1778
+ "step": 147500
1779
+ },
1780
+ {
1781
+ "epoch": 2.23,
1782
+ "learning_rate": 1.2845810111964654e-05,
1783
+ "loss": 0.6517,
1784
+ "step": 148000
1785
+ },
1786
+ {
1787
+ "epoch": 2.24,
1788
+ "learning_rate": 1.2720289200180752e-05,
1789
+ "loss": 0.6596,
1790
+ "step": 148500
1791
+ },
1792
+ {
1793
+ "epoch": 2.24,
1794
+ "learning_rate": 1.2594768288396847e-05,
1795
+ "loss": 0.6531,
1796
+ "step": 149000
1797
+ },
1798
+ {
1799
+ "epoch": 2.25,
1800
+ "learning_rate": 1.2469247376612943e-05,
1801
+ "loss": 0.6542,
1802
+ "step": 149500
1803
+ },
1804
+ {
1805
+ "epoch": 2.26,
1806
+ "learning_rate": 1.234372646482904e-05,
1807
+ "loss": 0.6586,
1808
+ "step": 150000
1809
+ },
1810
+ {
1811
+ "epoch": 2.27,
1812
+ "learning_rate": 1.2218205553045138e-05,
1813
+ "loss": 0.647,
1814
+ "step": 150500
1815
+ },
1816
+ {
1817
+ "epoch": 2.27,
1818
+ "learning_rate": 1.2092684641261234e-05,
1819
+ "loss": 0.6634,
1820
+ "step": 151000
1821
+ },
1822
+ {
1823
+ "epoch": 2.28,
1824
+ "learning_rate": 1.1967163729477332e-05,
1825
+ "loss": 0.65,
1826
+ "step": 151500
1827
+ },
1828
+ {
1829
+ "epoch": 2.29,
1830
+ "learning_rate": 1.1841642817693429e-05,
1831
+ "loss": 0.6467,
1832
+ "step": 152000
1833
+ },
1834
+ {
1835
+ "epoch": 2.3,
1836
+ "learning_rate": 1.1716121905909525e-05,
1837
+ "loss": 0.6595,
1838
+ "step": 152500
1839
+ },
1840
+ {
1841
+ "epoch": 2.3,
1842
+ "learning_rate": 1.1590600994125621e-05,
1843
+ "loss": 0.6564,
1844
+ "step": 153000
1845
+ },
1846
+ {
1847
+ "epoch": 2.31,
1848
+ "learning_rate": 1.1465080082341718e-05,
1849
+ "loss": 0.6583,
1850
+ "step": 153500
1851
+ },
1852
+ {
1853
+ "epoch": 2.32,
1854
+ "learning_rate": 1.1339559170557816e-05,
1855
+ "loss": 0.6623,
1856
+ "step": 154000
1857
+ },
1858
+ {
1859
+ "epoch": 2.33,
1860
+ "learning_rate": 1.1214038258773912e-05,
1861
+ "loss": 0.6557,
1862
+ "step": 154500
1863
+ },
1864
+ {
1865
+ "epoch": 2.33,
1866
+ "learning_rate": 1.1088517346990009e-05,
1867
+ "loss": 0.6523,
1868
+ "step": 155000
1869
+ },
1870
+ {
1871
+ "epoch": 2.34,
1872
+ "learning_rate": 1.0962996435206107e-05,
1873
+ "loss": 0.6707,
1874
+ "step": 155500
1875
+ },
1876
+ {
1877
+ "epoch": 2.35,
1878
+ "learning_rate": 1.0837475523422203e-05,
1879
+ "loss": 0.658,
1880
+ "step": 156000
1881
+ },
1882
+ {
1883
+ "epoch": 2.36,
1884
+ "learning_rate": 1.07119546116383e-05,
1885
+ "loss": 0.6381,
1886
+ "step": 156500
1887
+ },
1888
+ {
1889
+ "epoch": 2.36,
1890
+ "learning_rate": 1.0586433699854396e-05,
1891
+ "loss": 0.6633,
1892
+ "step": 157000
1893
+ },
1894
+ {
1895
+ "epoch": 2.37,
1896
+ "learning_rate": 1.0460912788070492e-05,
1897
+ "loss": 0.6568,
1898
+ "step": 157500
1899
+ },
1900
+ {
1901
+ "epoch": 2.38,
1902
+ "learning_rate": 1.033539187628659e-05,
1903
+ "loss": 0.6483,
1904
+ "step": 158000
1905
+ },
1906
+ {
1907
+ "epoch": 2.39,
1908
+ "learning_rate": 1.0209870964502687e-05,
1909
+ "loss": 0.6478,
1910
+ "step": 158500
1911
+ },
1912
+ {
1913
+ "epoch": 2.39,
1914
+ "learning_rate": 1.0084350052718783e-05,
1915
+ "loss": 0.6565,
1916
+ "step": 159000
1917
+ },
1918
+ {
1919
+ "epoch": 2.4,
1920
+ "learning_rate": 9.958829140934881e-06,
1921
+ "loss": 0.6487,
1922
+ "step": 159500
1923
+ },
1924
+ {
1925
+ "epoch": 2.41,
1926
+ "learning_rate": 9.833308229150977e-06,
1927
+ "loss": 0.6547,
1928
+ "step": 160000
1929
+ },
1930
+ {
1931
+ "epoch": 2.42,
1932
+ "learning_rate": 9.707787317367074e-06,
1933
+ "loss": 0.6511,
1934
+ "step": 160500
1935
+ },
1936
+ {
1937
+ "epoch": 2.43,
1938
+ "learning_rate": 9.58226640558317e-06,
1939
+ "loss": 0.6421,
1940
+ "step": 161000
1941
+ },
1942
+ {
1943
+ "epoch": 2.43,
1944
+ "learning_rate": 9.456745493799266e-06,
1945
+ "loss": 0.6417,
1946
+ "step": 161500
1947
+ },
1948
+ {
1949
+ "epoch": 2.44,
1950
+ "learning_rate": 9.331224582015364e-06,
1951
+ "loss": 0.6452,
1952
+ "step": 162000
1953
+ },
1954
+ {
1955
+ "epoch": 2.45,
1956
+ "learning_rate": 9.20570367023146e-06,
1957
+ "loss": 0.6525,
1958
+ "step": 162500
1959
+ },
1960
+ {
1961
+ "epoch": 2.46,
1962
+ "learning_rate": 9.080182758447557e-06,
1963
+ "loss": 0.6516,
1964
+ "step": 163000
1965
+ },
1966
+ {
1967
+ "epoch": 2.46,
1968
+ "learning_rate": 8.954661846663655e-06,
1969
+ "loss": 0.646,
1970
+ "step": 163500
1971
+ },
1972
+ {
1973
+ "epoch": 2.47,
1974
+ "learning_rate": 8.829140934879752e-06,
1975
+ "loss": 0.6428,
1976
+ "step": 164000
1977
+ },
1978
+ {
1979
+ "epoch": 2.48,
1980
+ "learning_rate": 8.703620023095848e-06,
1981
+ "loss": 0.6431,
1982
+ "step": 164500
1983
+ },
1984
+ {
1985
+ "epoch": 2.49,
1986
+ "learning_rate": 8.578099111311944e-06,
1987
+ "loss": 0.6477,
1988
+ "step": 165000
1989
+ },
1990
+ {
1991
+ "epoch": 2.49,
1992
+ "learning_rate": 8.45257819952804e-06,
1993
+ "loss": 0.6491,
1994
+ "step": 165500
1995
+ },
1996
+ {
1997
+ "epoch": 2.5,
1998
+ "learning_rate": 8.327057287744139e-06,
1999
+ "loss": 0.6482,
2000
+ "step": 166000
2001
+ },
2002
+ {
2003
+ "epoch": 2.51,
2004
+ "learning_rate": 8.201536375960235e-06,
2005
+ "loss": 0.6458,
2006
+ "step": 166500
2007
+ },
2008
+ {
2009
+ "epoch": 2.52,
2010
+ "learning_rate": 8.076015464176333e-06,
2011
+ "loss": 0.6483,
2012
+ "step": 167000
2013
+ },
2014
+ {
2015
+ "epoch": 2.52,
2016
+ "learning_rate": 7.95049455239243e-06,
2017
+ "loss": 0.6491,
2018
+ "step": 167500
2019
+ },
2020
+ {
2021
+ "epoch": 2.53,
2022
+ "learning_rate": 7.824973640608526e-06,
2023
+ "loss": 0.6408,
2024
+ "step": 168000
2025
+ },
2026
+ {
2027
+ "epoch": 2.54,
2028
+ "learning_rate": 7.699452728824622e-06,
2029
+ "loss": 0.6491,
2030
+ "step": 168500
2031
+ },
2032
+ {
2033
+ "epoch": 2.55,
2034
+ "learning_rate": 7.5739318170407196e-06,
2035
+ "loss": 0.6472,
2036
+ "step": 169000
2037
+ },
2038
+ {
2039
+ "epoch": 2.55,
2040
+ "learning_rate": 7.448410905256815e-06,
2041
+ "loss": 0.6422,
2042
+ "step": 169500
2043
+ },
2044
+ {
2045
+ "epoch": 2.56,
2046
+ "learning_rate": 7.322889993472913e-06,
2047
+ "loss": 0.6471,
2048
+ "step": 170000
2049
+ },
2050
+ {
2051
+ "epoch": 2.57,
2052
+ "learning_rate": 7.1973690816890095e-06,
2053
+ "loss": 0.6343,
2054
+ "step": 170500
2055
+ },
2056
+ {
2057
+ "epoch": 2.58,
2058
+ "learning_rate": 7.071848169905107e-06,
2059
+ "loss": 0.6446,
2060
+ "step": 171000
2061
+ },
2062
+ {
2063
+ "epoch": 2.58,
2064
+ "learning_rate": 6.946327258121203e-06,
2065
+ "loss": 0.6444,
2066
+ "step": 171500
2067
+ },
2068
+ {
2069
+ "epoch": 2.59,
2070
+ "learning_rate": 6.8208063463372995e-06,
2071
+ "loss": 0.6455,
2072
+ "step": 172000
2073
+ },
2074
+ {
2075
+ "epoch": 2.6,
2076
+ "learning_rate": 6.6952854345533975e-06,
2077
+ "loss": 0.6448,
2078
+ "step": 172500
2079
+ },
2080
+ {
2081
+ "epoch": 2.61,
2082
+ "learning_rate": 6.569764522769493e-06,
2083
+ "loss": 0.6367,
2084
+ "step": 173000
2085
+ },
2086
+ {
2087
+ "epoch": 2.61,
2088
+ "learning_rate": 6.444243610985591e-06,
2089
+ "loss": 0.6303,
2090
+ "step": 173500
2091
+ },
2092
+ {
2093
+ "epoch": 2.62,
2094
+ "learning_rate": 6.3187226992016875e-06,
2095
+ "loss": 0.6392,
2096
+ "step": 174000
2097
+ },
2098
+ {
2099
+ "epoch": 2.63,
2100
+ "learning_rate": 6.193201787417785e-06,
2101
+ "loss": 0.6296,
2102
+ "step": 174500
2103
+ },
2104
+ {
2105
+ "epoch": 2.64,
2106
+ "learning_rate": 6.06768087563388e-06,
2107
+ "loss": 0.6411,
2108
+ "step": 175000
2109
+ },
2110
+ {
2111
+ "epoch": 2.64,
2112
+ "learning_rate": 5.942159963849977e-06,
2113
+ "loss": 0.6432,
2114
+ "step": 175500
2115
+ },
2116
+ {
2117
+ "epoch": 2.65,
2118
+ "learning_rate": 5.816639052066075e-06,
2119
+ "loss": 0.6483,
2120
+ "step": 176000
2121
+ },
2122
+ {
2123
+ "epoch": 2.66,
2124
+ "learning_rate": 5.691118140282172e-06,
2125
+ "loss": 0.6355,
2126
+ "step": 176500
2127
+ },
2128
+ {
2129
+ "epoch": 2.67,
2130
+ "learning_rate": 5.565597228498268e-06,
2131
+ "loss": 0.6426,
2132
+ "step": 177000
2133
+ },
2134
+ {
2135
+ "epoch": 2.67,
2136
+ "learning_rate": 5.440076316714365e-06,
2137
+ "loss": 0.64,
2138
+ "step": 177500
2139
+ },
2140
+ {
2141
+ "epoch": 2.68,
2142
+ "learning_rate": 5.314555404930462e-06,
2143
+ "loss": 0.6434,
2144
+ "step": 178000
2145
+ },
2146
+ {
2147
+ "epoch": 2.69,
2148
+ "learning_rate": 5.189034493146558e-06,
2149
+ "loss": 0.6373,
2150
+ "step": 178500
2151
+ },
2152
+ {
2153
+ "epoch": 2.7,
2154
+ "learning_rate": 5.063513581362655e-06,
2155
+ "loss": 0.6281,
2156
+ "step": 179000
2157
+ },
2158
+ {
2159
+ "epoch": 2.7,
2160
+ "learning_rate": 4.937992669578752e-06,
2161
+ "loss": 0.6323,
2162
+ "step": 179500
2163
+ },
2164
+ {
2165
+ "epoch": 2.71,
2166
+ "learning_rate": 4.812471757794849e-06,
2167
+ "loss": 0.6334,
2168
+ "step": 180000
2169
+ },
2170
+ {
2171
+ "epoch": 2.72,
2172
+ "learning_rate": 4.686950846010945e-06,
2173
+ "loss": 0.6316,
2174
+ "step": 180500
2175
+ },
2176
+ {
2177
+ "epoch": 2.73,
2178
+ "learning_rate": 4.5614299342270425e-06,
2179
+ "loss": 0.6442,
2180
+ "step": 181000
2181
+ },
2182
+ {
2183
+ "epoch": 2.73,
2184
+ "learning_rate": 4.435909022443139e-06,
2185
+ "loss": 0.6429,
2186
+ "step": 181500
2187
+ },
2188
+ {
2189
+ "epoch": 2.74,
2190
+ "learning_rate": 4.310388110659236e-06,
2191
+ "loss": 0.6224,
2192
+ "step": 182000
2193
+ },
2194
+ {
2195
+ "epoch": 2.75,
2196
+ "learning_rate": 4.1848671988753325e-06,
2197
+ "loss": 0.6314,
2198
+ "step": 182500
2199
+ },
2200
+ {
2201
+ "epoch": 2.76,
2202
+ "learning_rate": 4.05934628709143e-06,
2203
+ "loss": 0.6576,
2204
+ "step": 183000
2205
+ },
2206
+ {
2207
+ "epoch": 2.76,
2208
+ "learning_rate": 3.933825375307527e-06,
2209
+ "loss": 0.6348,
2210
+ "step": 183500
2211
+ },
2212
+ {
2213
+ "epoch": 2.77,
2214
+ "learning_rate": 3.808304463523623e-06,
2215
+ "loss": 0.635,
2216
+ "step": 184000
2217
+ },
2218
+ {
2219
+ "epoch": 2.78,
2220
+ "learning_rate": 3.68278355173972e-06,
2221
+ "loss": 0.6332,
2222
+ "step": 184500
2223
+ },
2224
+ {
2225
+ "epoch": 2.79,
2226
+ "learning_rate": 3.557262639955817e-06,
2227
+ "loss": 0.6282,
2228
+ "step": 185000
2229
+ },
2230
+ {
2231
+ "epoch": 2.79,
2232
+ "learning_rate": 3.4317417281719137e-06,
2233
+ "loss": 0.6333,
2234
+ "step": 185500
2235
+ },
2236
+ {
2237
+ "epoch": 2.8,
2238
+ "learning_rate": 3.30622081638801e-06,
2239
+ "loss": 0.6345,
2240
+ "step": 186000
2241
+ },
2242
+ {
2243
+ "epoch": 2.81,
2244
+ "learning_rate": 3.1806999046041072e-06,
2245
+ "loss": 0.6408,
2246
+ "step": 186500
2247
+ },
2248
+ {
2249
+ "epoch": 2.82,
2250
+ "learning_rate": 3.055178992820204e-06,
2251
+ "loss": 0.6284,
2252
+ "step": 187000
2253
+ },
2254
+ {
2255
+ "epoch": 2.82,
2256
+ "learning_rate": 2.929658081036301e-06,
2257
+ "loss": 0.6332,
2258
+ "step": 187500
2259
+ },
2260
+ {
2261
+ "epoch": 2.83,
2262
+ "learning_rate": 2.8041371692523976e-06,
2263
+ "loss": 0.6361,
2264
+ "step": 188000
2265
+ },
2266
+ {
2267
+ "epoch": 2.84,
2268
+ "learning_rate": 2.6786162574684944e-06,
2269
+ "loss": 0.6307,
2270
+ "step": 188500
2271
+ },
2272
+ {
2273
+ "epoch": 2.85,
2274
+ "learning_rate": 2.553095345684591e-06,
2275
+ "loss": 0.6373,
2276
+ "step": 189000
2277
+ },
2278
+ {
2279
+ "epoch": 2.85,
2280
+ "learning_rate": 2.427574433900688e-06,
2281
+ "loss": 0.6421,
2282
+ "step": 189500
2283
+ },
2284
+ {
2285
+ "epoch": 2.86,
2286
+ "learning_rate": 2.3020535221167848e-06,
2287
+ "loss": 0.635,
2288
+ "step": 190000
2289
+ },
2290
+ {
2291
+ "epoch": 2.87,
2292
+ "learning_rate": 2.1765326103328816e-06,
2293
+ "loss": 0.6273,
2294
+ "step": 190500
2295
+ },
2296
+ {
2297
+ "epoch": 2.88,
2298
+ "learning_rate": 2.0510116985489784e-06,
2299
+ "loss": 0.6374,
2300
+ "step": 191000
2301
+ },
2302
+ {
2303
+ "epoch": 2.88,
2304
+ "learning_rate": 1.925490786765075e-06,
2305
+ "loss": 0.6333,
2306
+ "step": 191500
2307
+ },
2308
+ {
2309
+ "epoch": 2.89,
2310
+ "learning_rate": 1.799969874981172e-06,
2311
+ "loss": 0.6285,
2312
+ "step": 192000
2313
+ },
2314
+ {
2315
+ "epoch": 2.9,
2316
+ "learning_rate": 1.6744489631972687e-06,
2317
+ "loss": 0.6266,
2318
+ "step": 192500
2319
+ },
2320
+ {
2321
+ "epoch": 2.91,
2322
+ "learning_rate": 1.5489280514133655e-06,
2323
+ "loss": 0.6358,
2324
+ "step": 193000
2325
+ },
2326
+ {
2327
+ "epoch": 2.91,
2328
+ "learning_rate": 1.4234071396294623e-06,
2329
+ "loss": 0.6263,
2330
+ "step": 193500
2331
+ },
2332
+ {
2333
+ "epoch": 2.92,
2334
+ "learning_rate": 1.2978862278455591e-06,
2335
+ "loss": 0.6428,
2336
+ "step": 194000
2337
+ },
2338
+ {
2339
+ "epoch": 2.93,
2340
+ "learning_rate": 1.172365316061656e-06,
2341
+ "loss": 0.6295,
2342
+ "step": 194500
2343
+ },
2344
+ {
2345
+ "epoch": 2.94,
2346
+ "learning_rate": 1.0468444042777527e-06,
2347
+ "loss": 0.632,
2348
+ "step": 195000
2349
+ },
2350
+ {
2351
+ "epoch": 2.94,
2352
+ "learning_rate": 9.213234924938495e-07,
2353
+ "loss": 0.6203,
2354
+ "step": 195500
2355
+ },
2356
+ {
2357
+ "epoch": 2.95,
2358
+ "learning_rate": 7.958025807099463e-07,
2359
+ "loss": 0.6223,
2360
+ "step": 196000
2361
+ },
2362
+ {
2363
+ "epoch": 2.96,
2364
+ "learning_rate": 6.702816689260432e-07,
2365
+ "loss": 0.6341,
2366
+ "step": 196500
2367
+ },
2368
+ {
2369
+ "epoch": 2.97,
2370
+ "learning_rate": 5.4476075714214e-07,
2371
+ "loss": 0.628,
2372
+ "step": 197000
2373
+ },
2374
+ {
2375
+ "epoch": 2.97,
2376
+ "learning_rate": 4.192398453582367e-07,
2377
+ "loss": 0.6267,
2378
+ "step": 197500
2379
+ },
2380
+ {
2381
+ "epoch": 2.98,
2382
+ "learning_rate": 2.937189335743335e-07,
2383
+ "loss": 0.6196,
2384
+ "step": 198000
2385
+ },
2386
+ {
2387
+ "epoch": 2.99,
2388
+ "learning_rate": 1.681980217904303e-07,
2389
+ "loss": 0.6263,
2390
+ "step": 198500
2391
+ },
2392
+ {
2393
+ "epoch": 3.0,
2394
+ "learning_rate": 4.267711000652709e-08,
2395
+ "loss": 0.6211,
2396
+ "step": 199000
2397
+ },
2398
+ {
2399
+ "epoch": 3.0,
2400
+ "step": 199170,
2401
+ "total_flos": 2.210416975580037e+17,
2402
+ "train_loss": 0.564786915779832,
2403
+ "train_runtime": 39960.6259,
2404
+ "train_samples_per_second": 49.841,
2405
+ "train_steps_per_second": 4.984
2406
+ }
2407
+ ],
2408
+ "max_steps": 199170,
2409
+ "num_train_epochs": 3,
2410
+ "total_flos": 2.210416975580037e+17,
2411
+ "trial_name": null,
2412
+ "trial_params": null
2413
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1fa324e8ef9446afa2492bf677c6d6e74d0296c1e6b2aa557ac2e6837bb1a51
3
+ size 2799
vocab.json ADDED
The diff for this file is too large to render. See raw diff