add MT model
Browse files- all_results.json +15 -0
- config.json +38 -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 +883 -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": 11.5638,
|
| 4 |
+
"predict_gen_len": 49.476,
|
| 5 |
+
"predict_loss": 2.044224500656128,
|
| 6 |
+
"predict_runtime": 836.8467,
|
| 7 |
+
"predict_samples": 1500,
|
| 8 |
+
"predict_samples_per_second": 1.792,
|
| 9 |
+
"predict_steps_per_second": 0.179,
|
| 10 |
+
"train_loss": 1.0012687716917485,
|
| 11 |
+
"train_runtime": 19490.5067,
|
| 12 |
+
"train_samples": 239179,
|
| 13 |
+
"train_samples_per_second": 36.815,
|
| 14 |
+
"train_steps_per_second": 3.681
|
| 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": 128034,
|
| 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": 11.5638,
|
| 3 |
+
"predict_gen_len": 49.476,
|
| 4 |
+
"predict_loss": 2.044224500656128,
|
| 5 |
+
"predict_runtime": 836.8467,
|
| 6 |
+
"predict_samples": 1500,
|
| 7 |
+
"predict_samples_per_second": 1.792,
|
| 8 |
+
"predict_steps_per_second": 0.179
|
| 9 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:80c957a11f608ca2490310e0f0cdb005253c97c5e754d70778e78ac8df35203c
|
| 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": 1.0012687716917485,
|
| 4 |
+
"train_runtime": 19490.5067,
|
| 5 |
+
"train_samples": 239179,
|
| 6 |
+
"train_samples_per_second": 36.815,
|
| 7 |
+
"train_steps_per_second": 3.681
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,883 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"global_step": 71754,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 4.9651587367951616e-05,
|
| 13 |
+
"loss": 2.9853,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.04,
|
| 18 |
+
"learning_rate": 4.930317473590323e-05,
|
| 19 |
+
"loss": 2.2101,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.06,
|
| 24 |
+
"learning_rate": 4.895476210385484e-05,
|
| 25 |
+
"loss": 1.9807,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.08,
|
| 30 |
+
"learning_rate": 4.860634947180645e-05,
|
| 31 |
+
"loss": 1.8067,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.1,
|
| 36 |
+
"learning_rate": 4.825793683975807e-05,
|
| 37 |
+
"loss": 1.7071,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.13,
|
| 42 |
+
"learning_rate": 4.7909524207709675e-05,
|
| 43 |
+
"loss": 1.6716,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15,
|
| 48 |
+
"learning_rate": 4.756111157566129e-05,
|
| 49 |
+
"loss": 1.6077,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.17,
|
| 54 |
+
"learning_rate": 4.72126989436129e-05,
|
| 55 |
+
"loss": 1.5357,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.19,
|
| 60 |
+
"learning_rate": 4.6864286311564515e-05,
|
| 61 |
+
"loss": 1.4992,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.21,
|
| 66 |
+
"learning_rate": 4.651587367951613e-05,
|
| 67 |
+
"loss": 1.4536,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.23,
|
| 72 |
+
"learning_rate": 4.616746104746774e-05,
|
| 73 |
+
"loss": 1.4247,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.25,
|
| 78 |
+
"learning_rate": 4.5819048415419355e-05,
|
| 79 |
+
"loss": 1.4124,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.27,
|
| 84 |
+
"learning_rate": 4.547063578337096e-05,
|
| 85 |
+
"loss": 1.4125,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.29,
|
| 90 |
+
"learning_rate": 4.512222315132258e-05,
|
| 91 |
+
"loss": 1.358,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.31,
|
| 96 |
+
"learning_rate": 4.477381051927419e-05,
|
| 97 |
+
"loss": 1.3406,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.33,
|
| 102 |
+
"learning_rate": 4.44253978872258e-05,
|
| 103 |
+
"loss": 1.3251,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.36,
|
| 108 |
+
"learning_rate": 4.4076985255177414e-05,
|
| 109 |
+
"loss": 1.3072,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.38,
|
| 114 |
+
"learning_rate": 4.372857262312903e-05,
|
| 115 |
+
"loss": 1.3088,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.4,
|
| 120 |
+
"learning_rate": 4.338015999108064e-05,
|
| 121 |
+
"loss": 1.2935,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.42,
|
| 126 |
+
"learning_rate": 4.303174735903225e-05,
|
| 127 |
+
"loss": 1.2937,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.44,
|
| 132 |
+
"learning_rate": 4.268333472698387e-05,
|
| 133 |
+
"loss": 1.2535,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.46,
|
| 138 |
+
"learning_rate": 4.2334922094935473e-05,
|
| 139 |
+
"loss": 1.253,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.48,
|
| 144 |
+
"learning_rate": 4.198650946288709e-05,
|
| 145 |
+
"loss": 1.229,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.5,
|
| 150 |
+
"learning_rate": 4.16380968308387e-05,
|
| 151 |
+
"loss": 1.2517,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.52,
|
| 156 |
+
"learning_rate": 4.128968419879031e-05,
|
| 157 |
+
"loss": 1.2161,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.54,
|
| 162 |
+
"learning_rate": 4.0941271566741926e-05,
|
| 163 |
+
"loss": 1.2112,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.56,
|
| 168 |
+
"learning_rate": 4.059285893469354e-05,
|
| 169 |
+
"loss": 1.2249,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.59,
|
| 174 |
+
"learning_rate": 4.024444630264515e-05,
|
| 175 |
+
"loss": 1.1943,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.61,
|
| 180 |
+
"learning_rate": 3.989603367059676e-05,
|
| 181 |
+
"loss": 1.1875,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.63,
|
| 186 |
+
"learning_rate": 3.954762103854838e-05,
|
| 187 |
+
"loss": 1.1844,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.65,
|
| 192 |
+
"learning_rate": 3.9199208406499986e-05,
|
| 193 |
+
"loss": 1.1732,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.67,
|
| 198 |
+
"learning_rate": 3.88507957744516e-05,
|
| 199 |
+
"loss": 1.161,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.69,
|
| 204 |
+
"learning_rate": 3.850238314240321e-05,
|
| 205 |
+
"loss": 1.1668,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.71,
|
| 210 |
+
"learning_rate": 3.8153970510354826e-05,
|
| 211 |
+
"loss": 1.1866,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.73,
|
| 216 |
+
"learning_rate": 3.780555787830644e-05,
|
| 217 |
+
"loss": 1.1433,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.75,
|
| 222 |
+
"learning_rate": 3.745714524625805e-05,
|
| 223 |
+
"loss": 1.1662,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.77,
|
| 228 |
+
"learning_rate": 3.7108732614209665e-05,
|
| 229 |
+
"loss": 1.1432,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.79,
|
| 234 |
+
"learning_rate": 3.676031998216127e-05,
|
| 235 |
+
"loss": 1.113,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.82,
|
| 240 |
+
"learning_rate": 3.641190735011289e-05,
|
| 241 |
+
"loss": 1.1083,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.84,
|
| 246 |
+
"learning_rate": 3.60634947180645e-05,
|
| 247 |
+
"loss": 1.1354,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.86,
|
| 252 |
+
"learning_rate": 3.571508208601611e-05,
|
| 253 |
+
"loss": 1.1009,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.88,
|
| 258 |
+
"learning_rate": 3.5366669453967725e-05,
|
| 259 |
+
"loss": 1.0959,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.9,
|
| 264 |
+
"learning_rate": 3.501825682191934e-05,
|
| 265 |
+
"loss": 1.1142,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.92,
|
| 270 |
+
"learning_rate": 3.466984418987095e-05,
|
| 271 |
+
"loss": 1.0885,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.94,
|
| 276 |
+
"learning_rate": 3.4321431557822564e-05,
|
| 277 |
+
"loss": 1.1067,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.96,
|
| 282 |
+
"learning_rate": 3.397301892577418e-05,
|
| 283 |
+
"loss": 1.0969,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.98,
|
| 288 |
+
"learning_rate": 3.3624606293725784e-05,
|
| 289 |
+
"loss": 1.0864,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.0,
|
| 294 |
+
"learning_rate": 3.3276193661677404e-05,
|
| 295 |
+
"loss": 1.0521,
|
| 296 |
+
"step": 24000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 1.02,
|
| 300 |
+
"learning_rate": 3.292778102962901e-05,
|
| 301 |
+
"loss": 0.9488,
|
| 302 |
+
"step": 24500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.05,
|
| 306 |
+
"learning_rate": 3.2579368397580624e-05,
|
| 307 |
+
"loss": 0.9548,
|
| 308 |
+
"step": 25000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 1.07,
|
| 312 |
+
"learning_rate": 3.223095576553224e-05,
|
| 313 |
+
"loss": 0.9526,
|
| 314 |
+
"step": 25500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 1.09,
|
| 318 |
+
"learning_rate": 3.188254313348385e-05,
|
| 319 |
+
"loss": 0.9632,
|
| 320 |
+
"step": 26000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 1.11,
|
| 324 |
+
"learning_rate": 3.1534130501435464e-05,
|
| 325 |
+
"loss": 0.973,
|
| 326 |
+
"step": 26500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 1.13,
|
| 330 |
+
"learning_rate": 3.118571786938708e-05,
|
| 331 |
+
"loss": 0.9313,
|
| 332 |
+
"step": 27000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.15,
|
| 336 |
+
"learning_rate": 3.083730523733869e-05,
|
| 337 |
+
"loss": 0.9518,
|
| 338 |
+
"step": 27500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.17,
|
| 342 |
+
"learning_rate": 3.0488892605290297e-05,
|
| 343 |
+
"loss": 0.958,
|
| 344 |
+
"step": 28000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 1.19,
|
| 348 |
+
"learning_rate": 3.014047997324191e-05,
|
| 349 |
+
"loss": 0.9608,
|
| 350 |
+
"step": 28500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 1.21,
|
| 354 |
+
"learning_rate": 2.9792067341193523e-05,
|
| 355 |
+
"loss": 0.9505,
|
| 356 |
+
"step": 29000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 1.23,
|
| 360 |
+
"learning_rate": 2.9443654709145136e-05,
|
| 361 |
+
"loss": 0.9459,
|
| 362 |
+
"step": 29500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 1.25,
|
| 366 |
+
"learning_rate": 2.9095242077096746e-05,
|
| 367 |
+
"loss": 0.9714,
|
| 368 |
+
"step": 30000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.28,
|
| 372 |
+
"learning_rate": 2.8746829445048363e-05,
|
| 373 |
+
"loss": 0.9428,
|
| 374 |
+
"step": 30500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.3,
|
| 378 |
+
"learning_rate": 2.8398416812999973e-05,
|
| 379 |
+
"loss": 0.9437,
|
| 380 |
+
"step": 31000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.32,
|
| 384 |
+
"learning_rate": 2.8050004180951583e-05,
|
| 385 |
+
"loss": 0.9598,
|
| 386 |
+
"step": 31500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 1.34,
|
| 390 |
+
"learning_rate": 2.77015915489032e-05,
|
| 391 |
+
"loss": 0.9452,
|
| 392 |
+
"step": 32000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 1.36,
|
| 396 |
+
"learning_rate": 2.735317891685481e-05,
|
| 397 |
+
"loss": 0.9371,
|
| 398 |
+
"step": 32500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 1.38,
|
| 402 |
+
"learning_rate": 2.7004766284806422e-05,
|
| 403 |
+
"loss": 0.937,
|
| 404 |
+
"step": 33000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.4,
|
| 408 |
+
"learning_rate": 2.6656353652758035e-05,
|
| 409 |
+
"loss": 0.929,
|
| 410 |
+
"step": 33500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.42,
|
| 414 |
+
"learning_rate": 2.630794102070965e-05,
|
| 415 |
+
"loss": 0.9305,
|
| 416 |
+
"step": 34000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.44,
|
| 420 |
+
"learning_rate": 2.595952838866126e-05,
|
| 421 |
+
"loss": 0.9396,
|
| 422 |
+
"step": 34500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.46,
|
| 426 |
+
"learning_rate": 2.5611115756612875e-05,
|
| 427 |
+
"loss": 0.9248,
|
| 428 |
+
"step": 35000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 1.48,
|
| 432 |
+
"learning_rate": 2.5262703124564485e-05,
|
| 433 |
+
"loss": 0.9366,
|
| 434 |
+
"step": 35500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.51,
|
| 438 |
+
"learning_rate": 2.49142904925161e-05,
|
| 439 |
+
"loss": 0.9199,
|
| 440 |
+
"step": 36000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.53,
|
| 444 |
+
"learning_rate": 2.4565877860467708e-05,
|
| 445 |
+
"loss": 0.9239,
|
| 446 |
+
"step": 36500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.55,
|
| 450 |
+
"learning_rate": 2.421746522841932e-05,
|
| 451 |
+
"loss": 0.9351,
|
| 452 |
+
"step": 37000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.57,
|
| 456 |
+
"learning_rate": 2.3869052596370935e-05,
|
| 457 |
+
"loss": 0.916,
|
| 458 |
+
"step": 37500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.59,
|
| 462 |
+
"learning_rate": 2.3520639964322548e-05,
|
| 463 |
+
"loss": 0.919,
|
| 464 |
+
"step": 38000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.61,
|
| 468 |
+
"learning_rate": 2.317222733227416e-05,
|
| 469 |
+
"loss": 0.8994,
|
| 470 |
+
"step": 38500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.63,
|
| 474 |
+
"learning_rate": 2.2823814700225774e-05,
|
| 475 |
+
"loss": 0.9181,
|
| 476 |
+
"step": 39000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.65,
|
| 480 |
+
"learning_rate": 2.2475402068177384e-05,
|
| 481 |
+
"loss": 0.9098,
|
| 482 |
+
"step": 39500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.67,
|
| 486 |
+
"learning_rate": 2.2126989436128997e-05,
|
| 487 |
+
"loss": 0.908,
|
| 488 |
+
"step": 40000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.69,
|
| 492 |
+
"learning_rate": 2.177857680408061e-05,
|
| 493 |
+
"loss": 0.9131,
|
| 494 |
+
"step": 40500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.71,
|
| 498 |
+
"learning_rate": 2.143016417203222e-05,
|
| 499 |
+
"loss": 0.8964,
|
| 500 |
+
"step": 41000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.74,
|
| 504 |
+
"learning_rate": 2.1081751539983834e-05,
|
| 505 |
+
"loss": 0.9017,
|
| 506 |
+
"step": 41500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.76,
|
| 510 |
+
"learning_rate": 2.0733338907935447e-05,
|
| 511 |
+
"loss": 0.8914,
|
| 512 |
+
"step": 42000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.78,
|
| 516 |
+
"learning_rate": 2.038492627588706e-05,
|
| 517 |
+
"loss": 0.893,
|
| 518 |
+
"step": 42500
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.8,
|
| 522 |
+
"learning_rate": 2.0036513643838674e-05,
|
| 523 |
+
"loss": 0.8932,
|
| 524 |
+
"step": 43000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 1.82,
|
| 528 |
+
"learning_rate": 1.9688101011790283e-05,
|
| 529 |
+
"loss": 0.9003,
|
| 530 |
+
"step": 43500
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 1.84,
|
| 534 |
+
"learning_rate": 1.9339688379741897e-05,
|
| 535 |
+
"loss": 0.8818,
|
| 536 |
+
"step": 44000
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.86,
|
| 540 |
+
"learning_rate": 1.899127574769351e-05,
|
| 541 |
+
"loss": 0.8941,
|
| 542 |
+
"step": 44500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.88,
|
| 546 |
+
"learning_rate": 1.864286311564512e-05,
|
| 547 |
+
"loss": 0.8694,
|
| 548 |
+
"step": 45000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.9,
|
| 552 |
+
"learning_rate": 1.8294450483596733e-05,
|
| 553 |
+
"loss": 0.8887,
|
| 554 |
+
"step": 45500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 1.92,
|
| 558 |
+
"learning_rate": 1.7946037851548346e-05,
|
| 559 |
+
"loss": 0.8857,
|
| 560 |
+
"step": 46000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 1.94,
|
| 564 |
+
"learning_rate": 1.759762521949996e-05,
|
| 565 |
+
"loss": 0.8746,
|
| 566 |
+
"step": 46500
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 1.97,
|
| 570 |
+
"learning_rate": 1.7249212587451573e-05,
|
| 571 |
+
"loss": 0.8668,
|
| 572 |
+
"step": 47000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.99,
|
| 576 |
+
"learning_rate": 1.6900799955403186e-05,
|
| 577 |
+
"loss": 0.8817,
|
| 578 |
+
"step": 47500
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.01,
|
| 582 |
+
"learning_rate": 1.6552387323354796e-05,
|
| 583 |
+
"loss": 0.8397,
|
| 584 |
+
"step": 48000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 2.03,
|
| 588 |
+
"learning_rate": 1.620397469130641e-05,
|
| 589 |
+
"loss": 0.7568,
|
| 590 |
+
"step": 48500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 2.05,
|
| 594 |
+
"learning_rate": 1.5855562059258022e-05,
|
| 595 |
+
"loss": 0.7493,
|
| 596 |
+
"step": 49000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 2.07,
|
| 600 |
+
"learning_rate": 1.5507149427209632e-05,
|
| 601 |
+
"loss": 0.7577,
|
| 602 |
+
"step": 49500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 2.09,
|
| 606 |
+
"learning_rate": 1.5158736795161247e-05,
|
| 607 |
+
"loss": 0.7413,
|
| 608 |
+
"step": 50000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 2.11,
|
| 612 |
+
"learning_rate": 1.481032416311286e-05,
|
| 613 |
+
"loss": 0.7521,
|
| 614 |
+
"step": 50500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 2.13,
|
| 618 |
+
"learning_rate": 1.446191153106447e-05,
|
| 619 |
+
"loss": 0.7405,
|
| 620 |
+
"step": 51000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.15,
|
| 624 |
+
"learning_rate": 1.4113498899016083e-05,
|
| 625 |
+
"loss": 0.7474,
|
| 626 |
+
"step": 51500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 2.17,
|
| 630 |
+
"learning_rate": 1.3765086266967697e-05,
|
| 631 |
+
"loss": 0.7649,
|
| 632 |
+
"step": 52000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.19,
|
| 636 |
+
"learning_rate": 1.3416673634919308e-05,
|
| 637 |
+
"loss": 0.7551,
|
| 638 |
+
"step": 52500
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 2.22,
|
| 642 |
+
"learning_rate": 1.3068261002870921e-05,
|
| 643 |
+
"loss": 0.7677,
|
| 644 |
+
"step": 53000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 2.24,
|
| 648 |
+
"learning_rate": 1.2719848370822531e-05,
|
| 649 |
+
"loss": 0.7214,
|
| 650 |
+
"step": 53500
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 2.26,
|
| 654 |
+
"learning_rate": 1.2371435738774145e-05,
|
| 655 |
+
"loss": 0.7513,
|
| 656 |
+
"step": 54000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 2.28,
|
| 660 |
+
"learning_rate": 1.202302310672576e-05,
|
| 661 |
+
"loss": 0.7361,
|
| 662 |
+
"step": 54500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.3,
|
| 666 |
+
"learning_rate": 1.1674610474677371e-05,
|
| 667 |
+
"loss": 0.7561,
|
| 668 |
+
"step": 55000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 2.32,
|
| 672 |
+
"learning_rate": 1.1326197842628983e-05,
|
| 673 |
+
"loss": 0.7495,
|
| 674 |
+
"step": 55500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 2.34,
|
| 678 |
+
"learning_rate": 1.0977785210580594e-05,
|
| 679 |
+
"loss": 0.7528,
|
| 680 |
+
"step": 56000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 2.36,
|
| 684 |
+
"learning_rate": 1.0629372578532207e-05,
|
| 685 |
+
"loss": 0.7488,
|
| 686 |
+
"step": 56500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 2.38,
|
| 690 |
+
"learning_rate": 1.028095994648382e-05,
|
| 691 |
+
"loss": 0.7446,
|
| 692 |
+
"step": 57000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 2.4,
|
| 696 |
+
"learning_rate": 9.932547314435432e-06,
|
| 697 |
+
"loss": 0.7346,
|
| 698 |
+
"step": 57500
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 2.42,
|
| 702 |
+
"learning_rate": 9.584134682387045e-06,
|
| 703 |
+
"loss": 0.7392,
|
| 704 |
+
"step": 58000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.45,
|
| 708 |
+
"learning_rate": 9.235722050338657e-06,
|
| 709 |
+
"loss": 0.7414,
|
| 710 |
+
"step": 58500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 2.47,
|
| 714 |
+
"learning_rate": 8.88730941829027e-06,
|
| 715 |
+
"loss": 0.7366,
|
| 716 |
+
"step": 59000
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 2.49,
|
| 720 |
+
"learning_rate": 8.538896786241882e-06,
|
| 721 |
+
"loss": 0.7376,
|
| 722 |
+
"step": 59500
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 2.51,
|
| 726 |
+
"learning_rate": 8.190484154193495e-06,
|
| 727 |
+
"loss": 0.7466,
|
| 728 |
+
"step": 60000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 2.53,
|
| 732 |
+
"learning_rate": 7.842071522145107e-06,
|
| 733 |
+
"loss": 0.743,
|
| 734 |
+
"step": 60500
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 2.55,
|
| 738 |
+
"learning_rate": 7.493658890096719e-06,
|
| 739 |
+
"loss": 0.7447,
|
| 740 |
+
"step": 61000
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 2.57,
|
| 744 |
+
"learning_rate": 7.145246258048332e-06,
|
| 745 |
+
"loss": 0.7335,
|
| 746 |
+
"step": 61500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 2.59,
|
| 750 |
+
"learning_rate": 6.796833625999945e-06,
|
| 751 |
+
"loss": 0.7277,
|
| 752 |
+
"step": 62000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 2.61,
|
| 756 |
+
"learning_rate": 6.448420993951557e-06,
|
| 757 |
+
"loss": 0.745,
|
| 758 |
+
"step": 62500
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.63,
|
| 762 |
+
"learning_rate": 6.100008361903169e-06,
|
| 763 |
+
"loss": 0.7365,
|
| 764 |
+
"step": 63000
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 2.65,
|
| 768 |
+
"learning_rate": 5.751595729854782e-06,
|
| 769 |
+
"loss": 0.7334,
|
| 770 |
+
"step": 63500
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 2.68,
|
| 774 |
+
"learning_rate": 5.403183097806394e-06,
|
| 775 |
+
"loss": 0.7392,
|
| 776 |
+
"step": 64000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 2.7,
|
| 780 |
+
"learning_rate": 5.054770465758007e-06,
|
| 781 |
+
"loss": 0.7392,
|
| 782 |
+
"step": 64500
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 2.72,
|
| 786 |
+
"learning_rate": 4.70635783370962e-06,
|
| 787 |
+
"loss": 0.7318,
|
| 788 |
+
"step": 65000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 2.74,
|
| 792 |
+
"learning_rate": 4.357945201661231e-06,
|
| 793 |
+
"loss": 0.7208,
|
| 794 |
+
"step": 65500
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 2.76,
|
| 798 |
+
"learning_rate": 4.009532569612845e-06,
|
| 799 |
+
"loss": 0.7268,
|
| 800 |
+
"step": 66000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.78,
|
| 804 |
+
"learning_rate": 3.661119937564456e-06,
|
| 805 |
+
"loss": 0.734,
|
| 806 |
+
"step": 66500
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.8,
|
| 810 |
+
"learning_rate": 3.312707305516069e-06,
|
| 811 |
+
"loss": 0.7354,
|
| 812 |
+
"step": 67000
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 2.82,
|
| 816 |
+
"learning_rate": 2.9642946734676814e-06,
|
| 817 |
+
"loss": 0.733,
|
| 818 |
+
"step": 67500
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 2.84,
|
| 822 |
+
"learning_rate": 2.615882041419294e-06,
|
| 823 |
+
"loss": 0.7167,
|
| 824 |
+
"step": 68000
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 2.86,
|
| 828 |
+
"learning_rate": 2.267469409370906e-06,
|
| 829 |
+
"loss": 0.7181,
|
| 830 |
+
"step": 68500
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 2.88,
|
| 834 |
+
"learning_rate": 1.919056777322519e-06,
|
| 835 |
+
"loss": 0.7399,
|
| 836 |
+
"step": 69000
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 2.91,
|
| 840 |
+
"learning_rate": 1.5706441452741312e-06,
|
| 841 |
+
"loss": 0.7239,
|
| 842 |
+
"step": 69500
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.93,
|
| 846 |
+
"learning_rate": 1.2222315132257436e-06,
|
| 847 |
+
"loss": 0.7194,
|
| 848 |
+
"step": 70000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 2.95,
|
| 852 |
+
"learning_rate": 8.73818881177356e-07,
|
| 853 |
+
"loss": 0.7319,
|
| 854 |
+
"step": 70500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 2.97,
|
| 858 |
+
"learning_rate": 5.254062491289685e-07,
|
| 859 |
+
"loss": 0.7162,
|
| 860 |
+
"step": 71000
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 2.99,
|
| 864 |
+
"learning_rate": 1.7699361708058088e-07,
|
| 865 |
+
"loss": 0.7239,
|
| 866 |
+
"step": 71500
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 3.0,
|
| 870 |
+
"step": 71754,
|
| 871 |
+
"total_flos": 8.271547272486912e+16,
|
| 872 |
+
"train_loss": 1.0012687716917485,
|
| 873 |
+
"train_runtime": 19490.5067,
|
| 874 |
+
"train_samples_per_second": 36.815,
|
| 875 |
+
"train_steps_per_second": 3.681
|
| 876 |
+
}
|
| 877 |
+
],
|
| 878 |
+
"max_steps": 71754,
|
| 879 |
+
"num_train_epochs": 3,
|
| 880 |
+
"total_flos": 8.271547272486912e+16,
|
| 881 |
+
"trial_name": null,
|
| 882 |
+
"trial_params": null
|
| 883 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:93543e9ddf1801abacfd8d223703b3499e37b666ffd91d1e1bf041c531adafaa
|
| 3 |
+
size 2799
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|