add MT model
Browse files- all_results.json +15 -0
- config.json +38 -0
- eval_results.json +10 -0
- generated_predictions.txt +0 -0
- predict_results.json +9 -0
- pytorch_model.bin +3 -0
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- train_results.json +8 -0
- trainer_state.json +1699 -0
- training_args.bin +3 -0
- vocab.json +0 -0
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"predict_bleu": 11.8252,
|
| 4 |
+
"predict_gen_len": 71.3357,
|
| 5 |
+
"predict_loss": 1.8680634498596191,
|
| 6 |
+
"predict_runtime": 1006.451,
|
| 7 |
+
"predict_samples": 1558,
|
| 8 |
+
"predict_samples_per_second": 1.548,
|
| 9 |
+
"predict_steps_per_second": 0.155,
|
| 10 |
+
"train_loss": 0.8559671926554289,
|
| 11 |
+
"train_runtime": 48243.0463,
|
| 12 |
+
"train_samples": 466515,
|
| 13 |
+
"train_samples_per_second": 29.01,
|
| 14 |
+
"train_steps_per_second": 2.901
|
| 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": 128101,
|
| 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": "4.12.0",
|
| 36 |
+
"use_cache": true,
|
| 37 |
+
"vocab_size": 128112
|
| 38 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_bleu": 10.3486,
|
| 4 |
+
"eval_gen_len": 62.2768,
|
| 5 |
+
"eval_loss": 2.00561261177063,
|
| 6 |
+
"eval_runtime": 552.4579,
|
| 7 |
+
"eval_samples": 1391,
|
| 8 |
+
"eval_samples_per_second": 2.518,
|
| 9 |
+
"eval_steps_per_second": 0.127
|
| 10 |
+
}
|
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": 11.8252,
|
| 3 |
+
"predict_gen_len": 71.3357,
|
| 4 |
+
"predict_loss": 1.8680634498596191,
|
| 5 |
+
"predict_runtime": 1006.451,
|
| 6 |
+
"predict_samples": 1558,
|
| 7 |
+
"predict_samples_per_second": 1.548,
|
| 8 |
+
"predict_steps_per_second": 0.155
|
| 9 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8e945b58e0f5bc045f675af4186ca32cb62213b17688673f261640b913d10132
|
| 3 |
+
size 1935804116
|
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": 0.8559671926554289,
|
| 4 |
+
"train_runtime": 48243.0463,
|
| 5 |
+
"train_samples": 466515,
|
| 6 |
+
"train_samples_per_second": 29.01,
|
| 7 |
+
"train_steps_per_second": 2.901
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1699 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"global_step": 139956,
|
| 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": 4.9821372431335564e-05,
|
| 13 |
+
"loss": 3.0407,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 4.9642744862671126e-05,
|
| 19 |
+
"loss": 2.1865,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.03,
|
| 24 |
+
"learning_rate": 4.946411729400669e-05,
|
| 25 |
+
"loss": 1.9141,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.04,
|
| 30 |
+
"learning_rate": 4.9285489725342256e-05,
|
| 31 |
+
"loss": 1.766,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.05,
|
| 36 |
+
"learning_rate": 4.910686215667781e-05,
|
| 37 |
+
"loss": 1.6542,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.06,
|
| 42 |
+
"learning_rate": 4.892823458801338e-05,
|
| 43 |
+
"loss": 1.6047,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.08,
|
| 48 |
+
"learning_rate": 4.874960701934894e-05,
|
| 49 |
+
"loss": 1.53,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.09,
|
| 54 |
+
"learning_rate": 4.85709794506845e-05,
|
| 55 |
+
"loss": 1.4824,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.1,
|
| 60 |
+
"learning_rate": 4.8392351882020064e-05,
|
| 61 |
+
"loss": 1.4351,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.11,
|
| 66 |
+
"learning_rate": 4.821372431335563e-05,
|
| 67 |
+
"loss": 1.405,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.12,
|
| 72 |
+
"learning_rate": 4.8035096744691194e-05,
|
| 73 |
+
"loss": 1.3847,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.13,
|
| 78 |
+
"learning_rate": 4.7856469176026755e-05,
|
| 79 |
+
"loss": 1.3464,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.14,
|
| 84 |
+
"learning_rate": 4.767784160736232e-05,
|
| 85 |
+
"loss": 1.3163,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.15,
|
| 90 |
+
"learning_rate": 4.749921403869788e-05,
|
| 91 |
+
"loss": 1.3101,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.16,
|
| 96 |
+
"learning_rate": 4.732058647003344e-05,
|
| 97 |
+
"loss": 1.3054,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.17,
|
| 102 |
+
"learning_rate": 4.7141958901369e-05,
|
| 103 |
+
"loss": 1.2715,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.18,
|
| 108 |
+
"learning_rate": 4.696333133270457e-05,
|
| 109 |
+
"loss": 1.2606,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.19,
|
| 114 |
+
"learning_rate": 4.6784703764040125e-05,
|
| 115 |
+
"loss": 1.2423,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.2,
|
| 120 |
+
"learning_rate": 4.660607619537569e-05,
|
| 121 |
+
"loss": 1.2302,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.21,
|
| 126 |
+
"learning_rate": 4.6427448626711255e-05,
|
| 127 |
+
"loss": 1.2361,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.23,
|
| 132 |
+
"learning_rate": 4.6248821058046816e-05,
|
| 133 |
+
"loss": 1.2078,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.24,
|
| 138 |
+
"learning_rate": 4.607019348938238e-05,
|
| 139 |
+
"loss": 1.1992,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.25,
|
| 144 |
+
"learning_rate": 4.5891565920717946e-05,
|
| 145 |
+
"loss": 1.1775,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.26,
|
| 150 |
+
"learning_rate": 4.57129383520535e-05,
|
| 151 |
+
"loss": 1.1808,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.27,
|
| 156 |
+
"learning_rate": 4.553431078338907e-05,
|
| 157 |
+
"loss": 1.1644,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.28,
|
| 162 |
+
"learning_rate": 4.535568321472463e-05,
|
| 163 |
+
"loss": 1.1504,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.29,
|
| 168 |
+
"learning_rate": 4.517705564606019e-05,
|
| 169 |
+
"loss": 1.1445,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3,
|
| 174 |
+
"learning_rate": 4.4998428077395754e-05,
|
| 175 |
+
"loss": 1.1474,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.31,
|
| 180 |
+
"learning_rate": 4.4819800508731316e-05,
|
| 181 |
+
"loss": 1.1255,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.32,
|
| 186 |
+
"learning_rate": 4.464117294006688e-05,
|
| 187 |
+
"loss": 1.1065,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.33,
|
| 192 |
+
"learning_rate": 4.446254537140244e-05,
|
| 193 |
+
"loss": 1.1128,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.34,
|
| 198 |
+
"learning_rate": 4.428391780273801e-05,
|
| 199 |
+
"loss": 1.1114,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.35,
|
| 204 |
+
"learning_rate": 4.410529023407357e-05,
|
| 205 |
+
"loss": 1.106,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.36,
|
| 210 |
+
"learning_rate": 4.392666266540913e-05,
|
| 211 |
+
"loss": 1.1163,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.38,
|
| 216 |
+
"learning_rate": 4.374803509674469e-05,
|
| 217 |
+
"loss": 1.0843,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.39,
|
| 222 |
+
"learning_rate": 4.356940752808026e-05,
|
| 223 |
+
"loss": 1.099,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.4,
|
| 228 |
+
"learning_rate": 4.3390779959415816e-05,
|
| 229 |
+
"loss": 1.0785,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.41,
|
| 234 |
+
"learning_rate": 4.3212152390751384e-05,
|
| 235 |
+
"loss": 1.0842,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.42,
|
| 240 |
+
"learning_rate": 4.3033524822086946e-05,
|
| 241 |
+
"loss": 1.0623,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.43,
|
| 246 |
+
"learning_rate": 4.285489725342251e-05,
|
| 247 |
+
"loss": 1.0563,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.44,
|
| 252 |
+
"learning_rate": 4.267626968475807e-05,
|
| 253 |
+
"loss": 1.0568,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.45,
|
| 258 |
+
"learning_rate": 4.249764211609363e-05,
|
| 259 |
+
"loss": 1.0363,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.46,
|
| 264 |
+
"learning_rate": 4.231901454742919e-05,
|
| 265 |
+
"loss": 1.0361,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.47,
|
| 270 |
+
"learning_rate": 4.2140386978764754e-05,
|
| 271 |
+
"loss": 1.0522,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.48,
|
| 276 |
+
"learning_rate": 4.196175941010032e-05,
|
| 277 |
+
"loss": 1.0287,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.49,
|
| 282 |
+
"learning_rate": 4.178313184143588e-05,
|
| 283 |
+
"loss": 1.0361,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.5,
|
| 288 |
+
"learning_rate": 4.1604504272771445e-05,
|
| 289 |
+
"loss": 1.036,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.51,
|
| 294 |
+
"learning_rate": 4.142587670410701e-05,
|
| 295 |
+
"loss": 1.0123,
|
| 296 |
+
"step": 24000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.53,
|
| 300 |
+
"learning_rate": 4.124724913544257e-05,
|
| 301 |
+
"loss": 1.0257,
|
| 302 |
+
"step": 24500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.54,
|
| 306 |
+
"learning_rate": 4.106862156677813e-05,
|
| 307 |
+
"loss": 1.0405,
|
| 308 |
+
"step": 25000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.55,
|
| 312 |
+
"learning_rate": 4.08899939981137e-05,
|
| 313 |
+
"loss": 1.0048,
|
| 314 |
+
"step": 25500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.56,
|
| 318 |
+
"learning_rate": 4.071136642944925e-05,
|
| 319 |
+
"loss": 1.0212,
|
| 320 |
+
"step": 26000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.57,
|
| 324 |
+
"learning_rate": 4.053273886078482e-05,
|
| 325 |
+
"loss": 1.009,
|
| 326 |
+
"step": 26500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.58,
|
| 330 |
+
"learning_rate": 4.035411129212038e-05,
|
| 331 |
+
"loss": 0.9971,
|
| 332 |
+
"step": 27000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.59,
|
| 336 |
+
"learning_rate": 4.0175483723455945e-05,
|
| 337 |
+
"loss": 0.9901,
|
| 338 |
+
"step": 27500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.6,
|
| 342 |
+
"learning_rate": 3.9996856154791506e-05,
|
| 343 |
+
"loss": 1.0051,
|
| 344 |
+
"step": 28000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.61,
|
| 348 |
+
"learning_rate": 3.981822858612707e-05,
|
| 349 |
+
"loss": 1.0079,
|
| 350 |
+
"step": 28500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.62,
|
| 354 |
+
"learning_rate": 3.9639601017462636e-05,
|
| 355 |
+
"loss": 0.9895,
|
| 356 |
+
"step": 29000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.63,
|
| 360 |
+
"learning_rate": 3.946097344879819e-05,
|
| 361 |
+
"loss": 0.9962,
|
| 362 |
+
"step": 29500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.64,
|
| 366 |
+
"learning_rate": 3.928234588013376e-05,
|
| 367 |
+
"loss": 0.9996,
|
| 368 |
+
"step": 30000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.65,
|
| 372 |
+
"learning_rate": 3.910371831146932e-05,
|
| 373 |
+
"loss": 0.9601,
|
| 374 |
+
"step": 30500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.66,
|
| 378 |
+
"learning_rate": 3.892509074280488e-05,
|
| 379 |
+
"loss": 0.9753,
|
| 380 |
+
"step": 31000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.68,
|
| 384 |
+
"learning_rate": 3.8746463174140444e-05,
|
| 385 |
+
"loss": 0.9694,
|
| 386 |
+
"step": 31500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.69,
|
| 390 |
+
"learning_rate": 3.856783560547601e-05,
|
| 391 |
+
"loss": 0.9772,
|
| 392 |
+
"step": 32000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.7,
|
| 396 |
+
"learning_rate": 3.838920803681157e-05,
|
| 397 |
+
"loss": 0.9499,
|
| 398 |
+
"step": 32500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.71,
|
| 402 |
+
"learning_rate": 3.8210580468147136e-05,
|
| 403 |
+
"loss": 0.9681,
|
| 404 |
+
"step": 33000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.72,
|
| 408 |
+
"learning_rate": 3.80319528994827e-05,
|
| 409 |
+
"loss": 0.9411,
|
| 410 |
+
"step": 33500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.73,
|
| 414 |
+
"learning_rate": 3.785332533081826e-05,
|
| 415 |
+
"loss": 0.9631,
|
| 416 |
+
"step": 34000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.74,
|
| 420 |
+
"learning_rate": 3.767469776215382e-05,
|
| 421 |
+
"loss": 0.9602,
|
| 422 |
+
"step": 34500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.75,
|
| 426 |
+
"learning_rate": 3.749607019348938e-05,
|
| 427 |
+
"loss": 0.9639,
|
| 428 |
+
"step": 35000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.76,
|
| 432 |
+
"learning_rate": 3.7317442624824944e-05,
|
| 433 |
+
"loss": 0.9624,
|
| 434 |
+
"step": 35500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.77,
|
| 438 |
+
"learning_rate": 3.7138815056160506e-05,
|
| 439 |
+
"loss": 0.9317,
|
| 440 |
+
"step": 36000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.78,
|
| 444 |
+
"learning_rate": 3.6960187487496074e-05,
|
| 445 |
+
"loss": 0.9261,
|
| 446 |
+
"step": 36500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.79,
|
| 450 |
+
"learning_rate": 3.6781559918831636e-05,
|
| 451 |
+
"loss": 0.934,
|
| 452 |
+
"step": 37000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.8,
|
| 456 |
+
"learning_rate": 3.66029323501672e-05,
|
| 457 |
+
"loss": 0.9347,
|
| 458 |
+
"step": 37500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.81,
|
| 462 |
+
"learning_rate": 3.642430478150276e-05,
|
| 463 |
+
"loss": 0.9525,
|
| 464 |
+
"step": 38000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.83,
|
| 468 |
+
"learning_rate": 3.624567721283833e-05,
|
| 469 |
+
"loss": 0.9436,
|
| 470 |
+
"step": 38500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.84,
|
| 474 |
+
"learning_rate": 3.606704964417388e-05,
|
| 475 |
+
"loss": 0.9286,
|
| 476 |
+
"step": 39000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.85,
|
| 480 |
+
"learning_rate": 3.588842207550945e-05,
|
| 481 |
+
"loss": 0.9251,
|
| 482 |
+
"step": 39500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.86,
|
| 486 |
+
"learning_rate": 3.570979450684501e-05,
|
| 487 |
+
"loss": 0.9306,
|
| 488 |
+
"step": 40000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.87,
|
| 492 |
+
"learning_rate": 3.5531166938180574e-05,
|
| 493 |
+
"loss": 0.9318,
|
| 494 |
+
"step": 40500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.88,
|
| 498 |
+
"learning_rate": 3.5352539369516135e-05,
|
| 499 |
+
"loss": 0.9208,
|
| 500 |
+
"step": 41000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.89,
|
| 504 |
+
"learning_rate": 3.51739118008517e-05,
|
| 505 |
+
"loss": 0.9259,
|
| 506 |
+
"step": 41500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.9,
|
| 510 |
+
"learning_rate": 3.499528423218726e-05,
|
| 511 |
+
"loss": 0.9019,
|
| 512 |
+
"step": 42000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.91,
|
| 516 |
+
"learning_rate": 3.481665666352282e-05,
|
| 517 |
+
"loss": 0.9123,
|
| 518 |
+
"step": 42500
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.92,
|
| 522 |
+
"learning_rate": 3.463802909485839e-05,
|
| 523 |
+
"loss": 0.9135,
|
| 524 |
+
"step": 43000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.93,
|
| 528 |
+
"learning_rate": 3.445940152619394e-05,
|
| 529 |
+
"loss": 0.9216,
|
| 530 |
+
"step": 43500
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.94,
|
| 534 |
+
"learning_rate": 3.428077395752951e-05,
|
| 535 |
+
"loss": 0.9109,
|
| 536 |
+
"step": 44000
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.95,
|
| 540 |
+
"learning_rate": 3.410214638886507e-05,
|
| 541 |
+
"loss": 0.9165,
|
| 542 |
+
"step": 44500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.96,
|
| 546 |
+
"learning_rate": 3.3923518820200635e-05,
|
| 547 |
+
"loss": 0.8947,
|
| 548 |
+
"step": 45000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.98,
|
| 552 |
+
"learning_rate": 3.3744891251536196e-05,
|
| 553 |
+
"loss": 0.8987,
|
| 554 |
+
"step": 45500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.99,
|
| 558 |
+
"learning_rate": 3.3566263682871765e-05,
|
| 559 |
+
"loss": 0.9046,
|
| 560 |
+
"step": 46000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 1.0,
|
| 564 |
+
"learning_rate": 3.338763611420732e-05,
|
| 565 |
+
"loss": 0.9059,
|
| 566 |
+
"step": 46500
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 1.01,
|
| 570 |
+
"learning_rate": 3.320900854554289e-05,
|
| 571 |
+
"loss": 0.8544,
|
| 572 |
+
"step": 47000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.02,
|
| 576 |
+
"learning_rate": 3.303038097687845e-05,
|
| 577 |
+
"loss": 0.8138,
|
| 578 |
+
"step": 47500
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 1.03,
|
| 582 |
+
"learning_rate": 3.285175340821401e-05,
|
| 583 |
+
"loss": 0.8298,
|
| 584 |
+
"step": 48000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.04,
|
| 588 |
+
"learning_rate": 3.267312583954957e-05,
|
| 589 |
+
"loss": 0.8221,
|
| 590 |
+
"step": 48500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.05,
|
| 594 |
+
"learning_rate": 3.2494498270885134e-05,
|
| 595 |
+
"loss": 0.8253,
|
| 596 |
+
"step": 49000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 1.06,
|
| 600 |
+
"learning_rate": 3.23158707022207e-05,
|
| 601 |
+
"loss": 0.8283,
|
| 602 |
+
"step": 49500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.07,
|
| 606 |
+
"learning_rate": 3.213724313355626e-05,
|
| 607 |
+
"loss": 0.8233,
|
| 608 |
+
"step": 50000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.08,
|
| 612 |
+
"learning_rate": 3.1958615564891826e-05,
|
| 613 |
+
"loss": 0.8352,
|
| 614 |
+
"step": 50500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 1.09,
|
| 618 |
+
"learning_rate": 3.177998799622739e-05,
|
| 619 |
+
"loss": 0.8091,
|
| 620 |
+
"step": 51000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 1.1,
|
| 624 |
+
"learning_rate": 3.160136042756295e-05,
|
| 625 |
+
"loss": 0.8165,
|
| 626 |
+
"step": 51500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.11,
|
| 630 |
+
"learning_rate": 3.142273285889851e-05,
|
| 631 |
+
"loss": 0.8138,
|
| 632 |
+
"step": 52000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.13,
|
| 636 |
+
"learning_rate": 3.124410529023408e-05,
|
| 637 |
+
"loss": 0.813,
|
| 638 |
+
"step": 52500
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 1.14,
|
| 642 |
+
"learning_rate": 3.1065477721569634e-05,
|
| 643 |
+
"loss": 0.8255,
|
| 644 |
+
"step": 53000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 1.15,
|
| 648 |
+
"learning_rate": 3.08868501529052e-05,
|
| 649 |
+
"loss": 0.8225,
|
| 650 |
+
"step": 53500
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 1.16,
|
| 654 |
+
"learning_rate": 3.0708222584240764e-05,
|
| 655 |
+
"loss": 0.8126,
|
| 656 |
+
"step": 54000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 1.17,
|
| 660 |
+
"learning_rate": 3.0529595015576326e-05,
|
| 661 |
+
"loss": 0.8222,
|
| 662 |
+
"step": 54500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.18,
|
| 666 |
+
"learning_rate": 3.0350967446911887e-05,
|
| 667 |
+
"loss": 0.8229,
|
| 668 |
+
"step": 55000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.19,
|
| 672 |
+
"learning_rate": 3.0172339878247452e-05,
|
| 673 |
+
"loss": 0.8082,
|
| 674 |
+
"step": 55500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.2,
|
| 678 |
+
"learning_rate": 2.999371230958301e-05,
|
| 679 |
+
"loss": 0.8134,
|
| 680 |
+
"step": 56000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 1.21,
|
| 684 |
+
"learning_rate": 2.9815084740918575e-05,
|
| 685 |
+
"loss": 0.802,
|
| 686 |
+
"step": 56500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 1.22,
|
| 690 |
+
"learning_rate": 2.963645717225414e-05,
|
| 691 |
+
"loss": 0.8113,
|
| 692 |
+
"step": 57000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 1.23,
|
| 696 |
+
"learning_rate": 2.94578296035897e-05,
|
| 697 |
+
"loss": 0.8032,
|
| 698 |
+
"step": 57500
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 1.24,
|
| 702 |
+
"learning_rate": 2.9279202034925264e-05,
|
| 703 |
+
"loss": 0.815,
|
| 704 |
+
"step": 58000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.25,
|
| 708 |
+
"learning_rate": 2.910057446626083e-05,
|
| 709 |
+
"loss": 0.8118,
|
| 710 |
+
"step": 58500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.26,
|
| 714 |
+
"learning_rate": 2.892194689759639e-05,
|
| 715 |
+
"loss": 0.8076,
|
| 716 |
+
"step": 59000
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.28,
|
| 720 |
+
"learning_rate": 2.8743319328931952e-05,
|
| 721 |
+
"loss": 0.8103,
|
| 722 |
+
"step": 59500
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 1.29,
|
| 726 |
+
"learning_rate": 2.8564691760267513e-05,
|
| 727 |
+
"loss": 0.8171,
|
| 728 |
+
"step": 60000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 1.3,
|
| 732 |
+
"learning_rate": 2.838606419160308e-05,
|
| 733 |
+
"loss": 0.8126,
|
| 734 |
+
"step": 60500
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.31,
|
| 738 |
+
"learning_rate": 2.8207436622938637e-05,
|
| 739 |
+
"loss": 0.8106,
|
| 740 |
+
"step": 61000
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 1.32,
|
| 744 |
+
"learning_rate": 2.80288090542742e-05,
|
| 745 |
+
"loss": 0.803,
|
| 746 |
+
"step": 61500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.33,
|
| 750 |
+
"learning_rate": 2.7850181485609767e-05,
|
| 751 |
+
"loss": 0.7868,
|
| 752 |
+
"step": 62000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.34,
|
| 756 |
+
"learning_rate": 2.7671553916945325e-05,
|
| 757 |
+
"loss": 0.8008,
|
| 758 |
+
"step": 62500
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.35,
|
| 762 |
+
"learning_rate": 2.749292634828089e-05,
|
| 763 |
+
"loss": 0.8002,
|
| 764 |
+
"step": 63000
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 1.36,
|
| 768 |
+
"learning_rate": 2.7314298779616455e-05,
|
| 769 |
+
"loss": 0.8122,
|
| 770 |
+
"step": 63500
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 1.37,
|
| 774 |
+
"learning_rate": 2.7135671210952013e-05,
|
| 775 |
+
"loss": 0.7968,
|
| 776 |
+
"step": 64000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 1.38,
|
| 780 |
+
"learning_rate": 2.6957043642287578e-05,
|
| 781 |
+
"loss": 0.7949,
|
| 782 |
+
"step": 64500
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 1.39,
|
| 786 |
+
"learning_rate": 2.6778416073623143e-05,
|
| 787 |
+
"loss": 0.7983,
|
| 788 |
+
"step": 65000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 1.4,
|
| 792 |
+
"learning_rate": 2.65997885049587e-05,
|
| 793 |
+
"loss": 0.796,
|
| 794 |
+
"step": 65500
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.41,
|
| 798 |
+
"learning_rate": 2.6421160936294266e-05,
|
| 799 |
+
"loss": 0.7916,
|
| 800 |
+
"step": 66000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.43,
|
| 804 |
+
"learning_rate": 2.6242533367629828e-05,
|
| 805 |
+
"loss": 0.7874,
|
| 806 |
+
"step": 66500
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 1.44,
|
| 810 |
+
"learning_rate": 2.606390579896539e-05,
|
| 811 |
+
"loss": 0.79,
|
| 812 |
+
"step": 67000
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 1.45,
|
| 816 |
+
"learning_rate": 2.588527823030095e-05,
|
| 817 |
+
"loss": 0.8037,
|
| 818 |
+
"step": 67500
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 1.46,
|
| 822 |
+
"learning_rate": 2.5706650661636516e-05,
|
| 823 |
+
"loss": 0.8038,
|
| 824 |
+
"step": 68000
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 1.47,
|
| 828 |
+
"learning_rate": 2.5528023092972074e-05,
|
| 829 |
+
"loss": 0.7896,
|
| 830 |
+
"step": 68500
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 1.48,
|
| 834 |
+
"learning_rate": 2.534939552430764e-05,
|
| 835 |
+
"loss": 0.7999,
|
| 836 |
+
"step": 69000
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.49,
|
| 840 |
+
"learning_rate": 2.5170767955643204e-05,
|
| 841 |
+
"loss": 0.7911,
|
| 842 |
+
"step": 69500
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.5,
|
| 846 |
+
"learning_rate": 2.4992140386978766e-05,
|
| 847 |
+
"loss": 0.7965,
|
| 848 |
+
"step": 70000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 1.51,
|
| 852 |
+
"learning_rate": 2.4813512818314327e-05,
|
| 853 |
+
"loss": 0.7814,
|
| 854 |
+
"step": 70500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 1.52,
|
| 858 |
+
"learning_rate": 2.4634885249649892e-05,
|
| 859 |
+
"loss": 0.7759,
|
| 860 |
+
"step": 71000
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 1.53,
|
| 864 |
+
"learning_rate": 2.4456257680985454e-05,
|
| 865 |
+
"loss": 0.7906,
|
| 866 |
+
"step": 71500
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 1.54,
|
| 870 |
+
"learning_rate": 2.427763011232102e-05,
|
| 871 |
+
"loss": 0.7911,
|
| 872 |
+
"step": 72000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.55,
|
| 876 |
+
"learning_rate": 2.409900254365658e-05,
|
| 877 |
+
"loss": 0.7933,
|
| 878 |
+
"step": 72500
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.56,
|
| 882 |
+
"learning_rate": 2.3920374974992142e-05,
|
| 883 |
+
"loss": 0.772,
|
| 884 |
+
"step": 73000
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.58,
|
| 888 |
+
"learning_rate": 2.3741747406327704e-05,
|
| 889 |
+
"loss": 0.7895,
|
| 890 |
+
"step": 73500
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 1.59,
|
| 894 |
+
"learning_rate": 2.3563119837663265e-05,
|
| 895 |
+
"loss": 0.7757,
|
| 896 |
+
"step": 74000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 1.6,
|
| 900 |
+
"learning_rate": 2.3384492268998827e-05,
|
| 901 |
+
"loss": 0.7771,
|
| 902 |
+
"step": 74500
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 1.61,
|
| 906 |
+
"learning_rate": 2.3205864700334392e-05,
|
| 907 |
+
"loss": 0.7775,
|
| 908 |
+
"step": 75000
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 1.62,
|
| 912 |
+
"learning_rate": 2.3027237131669954e-05,
|
| 913 |
+
"loss": 0.7788,
|
| 914 |
+
"step": 75500
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 1.63,
|
| 918 |
+
"learning_rate": 2.2848609563005515e-05,
|
| 919 |
+
"loss": 0.773,
|
| 920 |
+
"step": 76000
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.64,
|
| 924 |
+
"learning_rate": 2.266998199434108e-05,
|
| 925 |
+
"loss": 0.7776,
|
| 926 |
+
"step": 76500
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.65,
|
| 930 |
+
"learning_rate": 2.2491354425676642e-05,
|
| 931 |
+
"loss": 0.7982,
|
| 932 |
+
"step": 77000
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 1.66,
|
| 936 |
+
"learning_rate": 2.2312726857012207e-05,
|
| 937 |
+
"loss": 0.777,
|
| 938 |
+
"step": 77500
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 1.67,
|
| 942 |
+
"learning_rate": 2.213409928834777e-05,
|
| 943 |
+
"loss": 0.7965,
|
| 944 |
+
"step": 78000
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 1.68,
|
| 948 |
+
"learning_rate": 2.195547171968333e-05,
|
| 949 |
+
"loss": 0.7835,
|
| 950 |
+
"step": 78500
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 1.69,
|
| 954 |
+
"learning_rate": 2.1776844151018895e-05,
|
| 955 |
+
"loss": 0.767,
|
| 956 |
+
"step": 79000
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 1.7,
|
| 960 |
+
"learning_rate": 2.1598216582354457e-05,
|
| 961 |
+
"loss": 0.7714,
|
| 962 |
+
"step": 79500
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.71,
|
| 966 |
+
"learning_rate": 2.1419589013690018e-05,
|
| 967 |
+
"loss": 0.7707,
|
| 968 |
+
"step": 80000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.73,
|
| 972 |
+
"learning_rate": 2.124096144502558e-05,
|
| 973 |
+
"loss": 0.7763,
|
| 974 |
+
"step": 80500
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 1.74,
|
| 978 |
+
"learning_rate": 2.106233387636114e-05,
|
| 979 |
+
"loss": 0.7559,
|
| 980 |
+
"step": 81000
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 1.75,
|
| 984 |
+
"learning_rate": 2.0883706307696703e-05,
|
| 985 |
+
"loss": 0.7597,
|
| 986 |
+
"step": 81500
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 1.76,
|
| 990 |
+
"learning_rate": 2.0705078739032268e-05,
|
| 991 |
+
"loss": 0.7717,
|
| 992 |
+
"step": 82000
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 1.77,
|
| 996 |
+
"learning_rate": 2.052645117036783e-05,
|
| 997 |
+
"loss": 0.7637,
|
| 998 |
+
"step": 82500
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 1.78,
|
| 1002 |
+
"learning_rate": 2.0347823601703395e-05,
|
| 1003 |
+
"loss": 0.7641,
|
| 1004 |
+
"step": 83000
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.79,
|
| 1008 |
+
"learning_rate": 2.0169196033038956e-05,
|
| 1009 |
+
"loss": 0.7778,
|
| 1010 |
+
"step": 83500
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.8,
|
| 1014 |
+
"learning_rate": 1.9990568464374518e-05,
|
| 1015 |
+
"loss": 0.7553,
|
| 1016 |
+
"step": 84000
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 1.81,
|
| 1020 |
+
"learning_rate": 1.9811940895710083e-05,
|
| 1021 |
+
"loss": 0.7672,
|
| 1022 |
+
"step": 84500
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 1.82,
|
| 1026 |
+
"learning_rate": 1.9633313327045644e-05,
|
| 1027 |
+
"loss": 0.7525,
|
| 1028 |
+
"step": 85000
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 1.83,
|
| 1032 |
+
"learning_rate": 1.9454685758381206e-05,
|
| 1033 |
+
"loss": 0.7748,
|
| 1034 |
+
"step": 85500
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 1.84,
|
| 1038 |
+
"learning_rate": 1.927605818971677e-05,
|
| 1039 |
+
"loss": 0.7632,
|
| 1040 |
+
"step": 86000
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 1.85,
|
| 1044 |
+
"learning_rate": 1.9097430621052333e-05,
|
| 1045 |
+
"loss": 0.7679,
|
| 1046 |
+
"step": 86500
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.86,
|
| 1050 |
+
"learning_rate": 1.8918803052387894e-05,
|
| 1051 |
+
"loss": 0.746,
|
| 1052 |
+
"step": 87000
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.88,
|
| 1056 |
+
"learning_rate": 1.8740175483723456e-05,
|
| 1057 |
+
"loss": 0.7637,
|
| 1058 |
+
"step": 87500
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 1.89,
|
| 1062 |
+
"learning_rate": 1.8561547915059017e-05,
|
| 1063 |
+
"loss": 0.7518,
|
| 1064 |
+
"step": 88000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 1.9,
|
| 1068 |
+
"learning_rate": 1.8382920346394582e-05,
|
| 1069 |
+
"loss": 0.7578,
|
| 1070 |
+
"step": 88500
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 1.91,
|
| 1074 |
+
"learning_rate": 1.8204292777730144e-05,
|
| 1075 |
+
"loss": 0.7611,
|
| 1076 |
+
"step": 89000
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 1.92,
|
| 1080 |
+
"learning_rate": 1.8025665209065706e-05,
|
| 1081 |
+
"loss": 0.7576,
|
| 1082 |
+
"step": 89500
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 1.93,
|
| 1086 |
+
"learning_rate": 1.784703764040127e-05,
|
| 1087 |
+
"loss": 0.7405,
|
| 1088 |
+
"step": 90000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.94,
|
| 1092 |
+
"learning_rate": 1.7668410071736832e-05,
|
| 1093 |
+
"loss": 0.7657,
|
| 1094 |
+
"step": 90500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 1.95,
|
| 1098 |
+
"learning_rate": 1.7489782503072394e-05,
|
| 1099 |
+
"loss": 0.7539,
|
| 1100 |
+
"step": 91000
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 1.96,
|
| 1104 |
+
"learning_rate": 1.731115493440796e-05,
|
| 1105 |
+
"loss": 0.7496,
|
| 1106 |
+
"step": 91500
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 1.97,
|
| 1110 |
+
"learning_rate": 1.713252736574352e-05,
|
| 1111 |
+
"loss": 0.7447,
|
| 1112 |
+
"step": 92000
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 1.98,
|
| 1116 |
+
"learning_rate": 1.6953899797079085e-05,
|
| 1117 |
+
"loss": 0.7582,
|
| 1118 |
+
"step": 92500
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 1.99,
|
| 1122 |
+
"learning_rate": 1.6775272228414647e-05,
|
| 1123 |
+
"loss": 0.747,
|
| 1124 |
+
"step": 93000
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 2.0,
|
| 1128 |
+
"learning_rate": 1.659664465975021e-05,
|
| 1129 |
+
"loss": 0.7202,
|
| 1130 |
+
"step": 93500
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 2.01,
|
| 1134 |
+
"learning_rate": 1.641801709108577e-05,
|
| 1135 |
+
"loss": 0.6645,
|
| 1136 |
+
"step": 94000
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.03,
|
| 1140 |
+
"learning_rate": 1.6239389522421332e-05,
|
| 1141 |
+
"loss": 0.6758,
|
| 1142 |
+
"step": 94500
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 2.04,
|
| 1146 |
+
"learning_rate": 1.6060761953756893e-05,
|
| 1147 |
+
"loss": 0.6758,
|
| 1148 |
+
"step": 95000
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 2.05,
|
| 1152 |
+
"learning_rate": 1.588213438509246e-05,
|
| 1153 |
+
"loss": 0.6657,
|
| 1154 |
+
"step": 95500
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 2.06,
|
| 1158 |
+
"learning_rate": 1.570350681642802e-05,
|
| 1159 |
+
"loss": 0.6665,
|
| 1160 |
+
"step": 96000
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 2.07,
|
| 1164 |
+
"learning_rate": 1.552487924776358e-05,
|
| 1165 |
+
"loss": 0.6546,
|
| 1166 |
+
"step": 96500
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 2.08,
|
| 1170 |
+
"learning_rate": 1.5346251679099147e-05,
|
| 1171 |
+
"loss": 0.6688,
|
| 1172 |
+
"step": 97000
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 2.09,
|
| 1176 |
+
"learning_rate": 1.5167624110434708e-05,
|
| 1177 |
+
"loss": 0.6745,
|
| 1178 |
+
"step": 97500
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 2.1,
|
| 1182 |
+
"learning_rate": 1.4988996541770273e-05,
|
| 1183 |
+
"loss": 0.6786,
|
| 1184 |
+
"step": 98000
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 2.11,
|
| 1188 |
+
"learning_rate": 1.4810368973105835e-05,
|
| 1189 |
+
"loss": 0.6826,
|
| 1190 |
+
"step": 98500
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 2.12,
|
| 1194 |
+
"learning_rate": 1.4631741404441396e-05,
|
| 1195 |
+
"loss": 0.6631,
|
| 1196 |
+
"step": 99000
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 2.13,
|
| 1200 |
+
"learning_rate": 1.445311383577696e-05,
|
| 1201 |
+
"loss": 0.6707,
|
| 1202 |
+
"step": 99500
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 2.14,
|
| 1206 |
+
"learning_rate": 1.4274486267112521e-05,
|
| 1207 |
+
"loss": 0.665,
|
| 1208 |
+
"step": 100000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 2.15,
|
| 1212 |
+
"learning_rate": 1.4095858698448083e-05,
|
| 1213 |
+
"loss": 0.6729,
|
| 1214 |
+
"step": 100500
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 2.16,
|
| 1218 |
+
"learning_rate": 1.3917231129783648e-05,
|
| 1219 |
+
"loss": 0.6633,
|
| 1220 |
+
"step": 101000
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 2.18,
|
| 1224 |
+
"learning_rate": 1.373860356111921e-05,
|
| 1225 |
+
"loss": 0.665,
|
| 1226 |
+
"step": 101500
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 2.19,
|
| 1230 |
+
"learning_rate": 1.3559975992454771e-05,
|
| 1231 |
+
"loss": 0.6712,
|
| 1232 |
+
"step": 102000
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 2.2,
|
| 1236 |
+
"learning_rate": 1.3381348423790336e-05,
|
| 1237 |
+
"loss": 0.665,
|
| 1238 |
+
"step": 102500
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 2.21,
|
| 1242 |
+
"learning_rate": 1.3202720855125898e-05,
|
| 1243 |
+
"loss": 0.671,
|
| 1244 |
+
"step": 103000
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 2.22,
|
| 1248 |
+
"learning_rate": 1.3024093286461461e-05,
|
| 1249 |
+
"loss": 0.6733,
|
| 1250 |
+
"step": 103500
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 2.23,
|
| 1254 |
+
"learning_rate": 1.2845465717797023e-05,
|
| 1255 |
+
"loss": 0.6646,
|
| 1256 |
+
"step": 104000
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 2.24,
|
| 1260 |
+
"learning_rate": 1.2666838149132584e-05,
|
| 1261 |
+
"loss": 0.6707,
|
| 1262 |
+
"step": 104500
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 2.25,
|
| 1266 |
+
"learning_rate": 1.2488210580468147e-05,
|
| 1267 |
+
"loss": 0.6591,
|
| 1268 |
+
"step": 105000
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 2.26,
|
| 1272 |
+
"learning_rate": 1.230958301180371e-05,
|
| 1273 |
+
"loss": 0.6626,
|
| 1274 |
+
"step": 105500
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 2.27,
|
| 1278 |
+
"learning_rate": 1.2130955443139274e-05,
|
| 1279 |
+
"loss": 0.6545,
|
| 1280 |
+
"step": 106000
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 2.28,
|
| 1284 |
+
"learning_rate": 1.1952327874474836e-05,
|
| 1285 |
+
"loss": 0.6652,
|
| 1286 |
+
"step": 106500
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 2.29,
|
| 1290 |
+
"learning_rate": 1.1773700305810397e-05,
|
| 1291 |
+
"loss": 0.6533,
|
| 1292 |
+
"step": 107000
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 2.3,
|
| 1296 |
+
"learning_rate": 1.159507273714596e-05,
|
| 1297 |
+
"loss": 0.6613,
|
| 1298 |
+
"step": 107500
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 2.32,
|
| 1302 |
+
"learning_rate": 1.1416445168481524e-05,
|
| 1303 |
+
"loss": 0.6618,
|
| 1304 |
+
"step": 108000
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 2.33,
|
| 1308 |
+
"learning_rate": 1.1237817599817085e-05,
|
| 1309 |
+
"loss": 0.659,
|
| 1310 |
+
"step": 108500
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 2.34,
|
| 1314 |
+
"learning_rate": 1.1059190031152649e-05,
|
| 1315 |
+
"loss": 0.6643,
|
| 1316 |
+
"step": 109000
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 2.35,
|
| 1320 |
+
"learning_rate": 1.0880562462488212e-05,
|
| 1321 |
+
"loss": 0.6622,
|
| 1322 |
+
"step": 109500
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 2.36,
|
| 1326 |
+
"learning_rate": 1.0701934893823774e-05,
|
| 1327 |
+
"loss": 0.6553,
|
| 1328 |
+
"step": 110000
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 2.37,
|
| 1332 |
+
"learning_rate": 1.0523307325159335e-05,
|
| 1333 |
+
"loss": 0.6566,
|
| 1334 |
+
"step": 110500
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 2.38,
|
| 1338 |
+
"learning_rate": 1.0344679756494899e-05,
|
| 1339 |
+
"loss": 0.6607,
|
| 1340 |
+
"step": 111000
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 2.39,
|
| 1344 |
+
"learning_rate": 1.0166052187830462e-05,
|
| 1345 |
+
"loss": 0.6465,
|
| 1346 |
+
"step": 111500
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 2.4,
|
| 1350 |
+
"learning_rate": 9.987424619166023e-06,
|
| 1351 |
+
"loss": 0.6546,
|
| 1352 |
+
"step": 112000
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 2.41,
|
| 1356 |
+
"learning_rate": 9.808797050501587e-06,
|
| 1357 |
+
"loss": 0.6674,
|
| 1358 |
+
"step": 112500
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 2.42,
|
| 1362 |
+
"learning_rate": 9.63016948183715e-06,
|
| 1363 |
+
"loss": 0.6641,
|
| 1364 |
+
"step": 113000
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 2.43,
|
| 1368 |
+
"learning_rate": 9.451541913172712e-06,
|
| 1369 |
+
"loss": 0.6569,
|
| 1370 |
+
"step": 113500
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 2.44,
|
| 1374 |
+
"learning_rate": 9.272914344508273e-06,
|
| 1375 |
+
"loss": 0.6515,
|
| 1376 |
+
"step": 114000
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 2.45,
|
| 1380 |
+
"learning_rate": 9.094286775843837e-06,
|
| 1381 |
+
"loss": 0.6528,
|
| 1382 |
+
"step": 114500
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 2.47,
|
| 1386 |
+
"learning_rate": 8.9156592071794e-06,
|
| 1387 |
+
"loss": 0.6386,
|
| 1388 |
+
"step": 115000
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 2.48,
|
| 1392 |
+
"learning_rate": 8.737031638514963e-06,
|
| 1393 |
+
"loss": 0.655,
|
| 1394 |
+
"step": 115500
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 2.49,
|
| 1398 |
+
"learning_rate": 8.558404069850525e-06,
|
| 1399 |
+
"loss": 0.6534,
|
| 1400 |
+
"step": 116000
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 2.5,
|
| 1404 |
+
"learning_rate": 8.379776501186088e-06,
|
| 1405 |
+
"loss": 0.6557,
|
| 1406 |
+
"step": 116500
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 2.51,
|
| 1410 |
+
"learning_rate": 8.20114893252165e-06,
|
| 1411 |
+
"loss": 0.661,
|
| 1412 |
+
"step": 117000
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 2.52,
|
| 1416 |
+
"learning_rate": 8.022521363857213e-06,
|
| 1417 |
+
"loss": 0.6463,
|
| 1418 |
+
"step": 117500
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 2.53,
|
| 1422 |
+
"learning_rate": 7.843893795192775e-06,
|
| 1423 |
+
"loss": 0.6652,
|
| 1424 |
+
"step": 118000
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 2.54,
|
| 1428 |
+
"learning_rate": 7.665266226528338e-06,
|
| 1429 |
+
"loss": 0.6587,
|
| 1430 |
+
"step": 118500
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 2.55,
|
| 1434 |
+
"learning_rate": 7.486638657863901e-06,
|
| 1435 |
+
"loss": 0.6448,
|
| 1436 |
+
"step": 119000
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 2.56,
|
| 1440 |
+
"learning_rate": 7.308011089199463e-06,
|
| 1441 |
+
"loss": 0.6624,
|
| 1442 |
+
"step": 119500
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 2.57,
|
| 1446 |
+
"learning_rate": 7.129383520535025e-06,
|
| 1447 |
+
"loss": 0.655,
|
| 1448 |
+
"step": 120000
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 2.58,
|
| 1452 |
+
"learning_rate": 6.9507559518705885e-06,
|
| 1453 |
+
"loss": 0.6524,
|
| 1454 |
+
"step": 120500
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 2.59,
|
| 1458 |
+
"learning_rate": 6.772128383206151e-06,
|
| 1459 |
+
"loss": 0.6535,
|
| 1460 |
+
"step": 121000
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 2.6,
|
| 1464 |
+
"learning_rate": 6.5935008145417126e-06,
|
| 1465 |
+
"loss": 0.6509,
|
| 1466 |
+
"step": 121500
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 2.62,
|
| 1470 |
+
"learning_rate": 6.414873245877276e-06,
|
| 1471 |
+
"loss": 0.6528,
|
| 1472 |
+
"step": 122000
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 2.63,
|
| 1476 |
+
"learning_rate": 6.236245677212839e-06,
|
| 1477 |
+
"loss": 0.6625,
|
| 1478 |
+
"step": 122500
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 2.64,
|
| 1482 |
+
"learning_rate": 6.057618108548401e-06,
|
| 1483 |
+
"loss": 0.6533,
|
| 1484 |
+
"step": 123000
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 2.65,
|
| 1488 |
+
"learning_rate": 5.878990539883964e-06,
|
| 1489 |
+
"loss": 0.6405,
|
| 1490 |
+
"step": 123500
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 2.66,
|
| 1494 |
+
"learning_rate": 5.7003629712195265e-06,
|
| 1495 |
+
"loss": 0.6562,
|
| 1496 |
+
"step": 124000
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 2.67,
|
| 1500 |
+
"learning_rate": 5.521735402555089e-06,
|
| 1501 |
+
"loss": 0.6458,
|
| 1502 |
+
"step": 124500
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 2.68,
|
| 1506 |
+
"learning_rate": 5.343107833890651e-06,
|
| 1507 |
+
"loss": 0.6448,
|
| 1508 |
+
"step": 125000
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 2.69,
|
| 1512 |
+
"learning_rate": 5.164480265226214e-06,
|
| 1513 |
+
"loss": 0.6486,
|
| 1514 |
+
"step": 125500
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 2.7,
|
| 1518 |
+
"learning_rate": 4.985852696561777e-06,
|
| 1519 |
+
"loss": 0.6559,
|
| 1520 |
+
"step": 126000
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 2.71,
|
| 1524 |
+
"learning_rate": 4.807225127897339e-06,
|
| 1525 |
+
"loss": 0.6329,
|
| 1526 |
+
"step": 126500
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 2.72,
|
| 1530 |
+
"learning_rate": 4.628597559232902e-06,
|
| 1531 |
+
"loss": 0.6447,
|
| 1532 |
+
"step": 127000
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 2.73,
|
| 1536 |
+
"learning_rate": 4.4499699905684645e-06,
|
| 1537 |
+
"loss": 0.6551,
|
| 1538 |
+
"step": 127500
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 2.74,
|
| 1542 |
+
"learning_rate": 4.271342421904028e-06,
|
| 1543 |
+
"loss": 0.6441,
|
| 1544 |
+
"step": 128000
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 2.75,
|
| 1548 |
+
"learning_rate": 4.092714853239589e-06,
|
| 1549 |
+
"loss": 0.6481,
|
| 1550 |
+
"step": 128500
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 2.77,
|
| 1554 |
+
"learning_rate": 3.914087284575153e-06,
|
| 1555 |
+
"loss": 0.6405,
|
| 1556 |
+
"step": 129000
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 2.78,
|
| 1560 |
+
"learning_rate": 3.7354597159107147e-06,
|
| 1561 |
+
"loss": 0.6376,
|
| 1562 |
+
"step": 129500
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 2.79,
|
| 1566 |
+
"learning_rate": 3.556832147246278e-06,
|
| 1567 |
+
"loss": 0.6509,
|
| 1568 |
+
"step": 130000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 2.8,
|
| 1572 |
+
"learning_rate": 3.37820457858184e-06,
|
| 1573 |
+
"loss": 0.6321,
|
| 1574 |
+
"step": 130500
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 2.81,
|
| 1578 |
+
"learning_rate": 3.1995770099174025e-06,
|
| 1579 |
+
"loss": 0.6372,
|
| 1580 |
+
"step": 131000
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 2.82,
|
| 1584 |
+
"learning_rate": 3.0209494412529654e-06,
|
| 1585 |
+
"loss": 0.6348,
|
| 1586 |
+
"step": 131500
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 2.83,
|
| 1590 |
+
"learning_rate": 2.8423218725885283e-06,
|
| 1591 |
+
"loss": 0.6541,
|
| 1592 |
+
"step": 132000
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 2.84,
|
| 1596 |
+
"learning_rate": 2.6636943039240907e-06,
|
| 1597 |
+
"loss": 0.6462,
|
| 1598 |
+
"step": 132500
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 2.85,
|
| 1602 |
+
"learning_rate": 2.485066735259653e-06,
|
| 1603 |
+
"loss": 0.6451,
|
| 1604 |
+
"step": 133000
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 2.86,
|
| 1608 |
+
"learning_rate": 2.306439166595216e-06,
|
| 1609 |
+
"loss": 0.6398,
|
| 1610 |
+
"step": 133500
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 2.87,
|
| 1614 |
+
"learning_rate": 2.127811597930778e-06,
|
| 1615 |
+
"loss": 0.6435,
|
| 1616 |
+
"step": 134000
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 2.88,
|
| 1620 |
+
"learning_rate": 1.949184029266341e-06,
|
| 1621 |
+
"loss": 0.6371,
|
| 1622 |
+
"step": 134500
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 2.89,
|
| 1626 |
+
"learning_rate": 1.7705564606019034e-06,
|
| 1627 |
+
"loss": 0.6291,
|
| 1628 |
+
"step": 135000
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 2.9,
|
| 1632 |
+
"learning_rate": 1.591928891937466e-06,
|
| 1633 |
+
"loss": 0.6437,
|
| 1634 |
+
"step": 135500
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 2.92,
|
| 1638 |
+
"learning_rate": 1.4133013232730287e-06,
|
| 1639 |
+
"loss": 0.6429,
|
| 1640 |
+
"step": 136000
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 2.93,
|
| 1644 |
+
"learning_rate": 1.2346737546085914e-06,
|
| 1645 |
+
"loss": 0.6381,
|
| 1646 |
+
"step": 136500
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 2.94,
|
| 1650 |
+
"learning_rate": 1.056046185944154e-06,
|
| 1651 |
+
"loss": 0.6388,
|
| 1652 |
+
"step": 137000
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 2.95,
|
| 1656 |
+
"learning_rate": 8.774186172797166e-07,
|
| 1657 |
+
"loss": 0.6388,
|
| 1658 |
+
"step": 137500
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 2.96,
|
| 1662 |
+
"learning_rate": 6.987910486152791e-07,
|
| 1663 |
+
"loss": 0.639,
|
| 1664 |
+
"step": 138000
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 2.97,
|
| 1668 |
+
"learning_rate": 5.201634799508417e-07,
|
| 1669 |
+
"loss": 0.6275,
|
| 1670 |
+
"step": 138500
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 2.98,
|
| 1674 |
+
"learning_rate": 3.415359112864043e-07,
|
| 1675 |
+
"loss": 0.6425,
|
| 1676 |
+
"step": 139000
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 2.99,
|
| 1680 |
+
"learning_rate": 1.629083426219669e-07,
|
| 1681 |
+
"loss": 0.6301,
|
| 1682 |
+
"step": 139500
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 3.0,
|
| 1686 |
+
"step": 139956,
|
| 1687 |
+
"total_flos": 1.4238860552921088e+17,
|
| 1688 |
+
"train_loss": 0.8559671926554289,
|
| 1689 |
+
"train_runtime": 48243.0463,
|
| 1690 |
+
"train_samples_per_second": 29.01,
|
| 1691 |
+
"train_steps_per_second": 2.901
|
| 1692 |
+
}
|
| 1693 |
+
],
|
| 1694 |
+
"max_steps": 139956,
|
| 1695 |
+
"num_train_epochs": 3,
|
| 1696 |
+
"total_flos": 1.4238860552921088e+17,
|
| 1697 |
+
"trial_name": null,
|
| 1698 |
+
"trial_params": null
|
| 1699 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4087bbcb1791d31ba96c441b8f7b31f0e6475368e59b18afee6bafd1784fc537
|
| 3 |
+
size 2927
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|