Davide Montebovi
commited on
Commit
·
4455348
1
Parent(s):
acced76
first version
Browse files- all_results.json +21 -0
- cfg.json +31 -0
- config.json +30 -0
- eval_results.json +15 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- train_results.json +9 -0
- trainer_state.json +1792 -0
- training_args.bin +3 -0
all_results.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_exact_match": 5.6,
|
| 4 |
+
"eval_f1": 22.9652,
|
| 5 |
+
"eval_gen_len": 5.812,
|
| 6 |
+
"eval_loss": 0.544481098651886,
|
| 7 |
+
"eval_rouge1": 28.2152,
|
| 8 |
+
"eval_rouge2": 12.4988,
|
| 9 |
+
"eval_rougeL": 28.1392,
|
| 10 |
+
"eval_rougeLsum": 28.1588,
|
| 11 |
+
"eval_runtime": 76.7747,
|
| 12 |
+
"eval_samples": 1000,
|
| 13 |
+
"eval_samples_per_second": 13.025,
|
| 14 |
+
"eval_steps_per_second": 4.35,
|
| 15 |
+
"total_flos": 1.4035529182858445e+17,
|
| 16 |
+
"train_loss": 0.5681342689636171,
|
| 17 |
+
"train_runtime": 25047.2432,
|
| 18 |
+
"train_samples": 91106,
|
| 19 |
+
"train_samples_per_second": 10.912,
|
| 20 |
+
"train_steps_per_second": 3.637
|
| 21 |
+
}
|
cfg.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_name_or_path": "gsarti/it5-base",
|
| 3 |
+
"tokenizer_name": "gsarti/it5-base",
|
| 4 |
+
"dataset_name": "/content/drive/MyDrive/_colab/squad_code/my_squad_it",
|
| 5 |
+
"do_train": true,
|
| 6 |
+
"do_eval":false,
|
| 7 |
+
|
| 8 |
+
"push_to_hub" : false,
|
| 9 |
+
"per_device_train_batch_size": 3,
|
| 10 |
+
"per_device_eval_batch_size": 3,
|
| 11 |
+
"learning_rate": 3e-5,
|
| 12 |
+
"num_train_epochs": 3,
|
| 13 |
+
"max_seq_length": 384,
|
| 14 |
+
"doc_stride": 128,
|
| 15 |
+
"max_answer_length":30,
|
| 16 |
+
"n_best_size":5,
|
| 17 |
+
"save_steps": 500000,
|
| 18 |
+
"save_total_limit": 3,
|
| 19 |
+
"save_strategy":"steps",
|
| 20 |
+
"context_column": "context",
|
| 21 |
+
"question_column": "question",
|
| 22 |
+
"answer_column": "answers",
|
| 23 |
+
"output_dir": "./tmp/quanIta3-t5/",
|
| 24 |
+
|
| 25 |
+
"version_2_with_negative": true,
|
| 26 |
+
"eval_steps": 2000,
|
| 27 |
+
"evaluation_strategy": "steps",
|
| 28 |
+
"predict_with_generate":true,
|
| 29 |
+
"max_eval_samples": 1000
|
| 30 |
+
}
|
| 31 |
+
|
config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "gsarti/it5-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"T5ForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"d_ff": 2048,
|
| 7 |
+
"d_kv": 64,
|
| 8 |
+
"d_model": 768,
|
| 9 |
+
"decoder_start_token_id": 0,
|
| 10 |
+
"dropout_rate": 0.1,
|
| 11 |
+
"eos_token_id": 1,
|
| 12 |
+
"feed_forward_proj": "gated-gelu",
|
| 13 |
+
"gradient_checkpointing": false,
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"is_encoder_decoder": true,
|
| 16 |
+
"layer_norm_epsilon": 1e-06,
|
| 17 |
+
"model_type": "t5",
|
| 18 |
+
"num_decoder_layers": 12,
|
| 19 |
+
"num_heads": 12,
|
| 20 |
+
"num_layers": 12,
|
| 21 |
+
"output_past": true,
|
| 22 |
+
"pad_token_id": 0,
|
| 23 |
+
"relative_attention_max_distance": 128,
|
| 24 |
+
"relative_attention_num_buckets": 32,
|
| 25 |
+
"tie_word_embeddings": false,
|
| 26 |
+
"torch_dtype": "float32",
|
| 27 |
+
"transformers_version": "4.18.0",
|
| 28 |
+
"use_cache": true,
|
| 29 |
+
"vocab_size": 32103
|
| 30 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_exact_match": 5.6,
|
| 4 |
+
"eval_f1": 22.9652,
|
| 5 |
+
"eval_gen_len": 5.812,
|
| 6 |
+
"eval_loss": 0.544481098651886,
|
| 7 |
+
"eval_rouge1": 28.2152,
|
| 8 |
+
"eval_rouge2": 12.4988,
|
| 9 |
+
"eval_rougeL": 28.1392,
|
| 10 |
+
"eval_rougeLsum": 28.1588,
|
| 11 |
+
"eval_runtime": 76.7747,
|
| 12 |
+
"eval_samples": 1000,
|
| 13 |
+
"eval_samples_per_second": 13.025,
|
| 14 |
+
"eval_steps_per_second": 4.35
|
| 15 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:845f4527381cc5ac7c25e72af2a57905a8920cd81b681fe7be09df3034af7c00
|
| 3 |
+
size 990253005
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"]}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 100, "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"], "special_tokens_map_file": null, "name_or_path": "gsarti/it5-base", "tokenizer_class": "T5Tokenizer"}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 1.4035529182858445e+17,
|
| 4 |
+
"train_loss": 0.5681342689636171,
|
| 5 |
+
"train_runtime": 25047.2432,
|
| 6 |
+
"train_samples": 91106,
|
| 7 |
+
"train_samples_per_second": 10.912,
|
| 8 |
+
"train_steps_per_second": 3.637
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1792 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"global_step": 91107,
|
| 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": 2.983535842470941e-05,
|
| 13 |
+
"loss": 4.3991,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.03,
|
| 18 |
+
"learning_rate": 2.9670716849418816e-05,
|
| 19 |
+
"loss": 1.3497,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.05,
|
| 24 |
+
"learning_rate": 2.9506075274128224e-05,
|
| 25 |
+
"loss": 1.0917,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.07,
|
| 30 |
+
"learning_rate": 2.934143369883763e-05,
|
| 31 |
+
"loss": 0.8889,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.07,
|
| 36 |
+
"eval_exact_match": 18.4,
|
| 37 |
+
"eval_f1": 18.8036,
|
| 38 |
+
"eval_gen_len": 5.368,
|
| 39 |
+
"eval_loss": 0.6919174194335938,
|
| 40 |
+
"eval_rouge1": 18.1786,
|
| 41 |
+
"eval_rouge2": 8.6387,
|
| 42 |
+
"eval_rougeL": 18.1534,
|
| 43 |
+
"eval_rougeLsum": 18.0938,
|
| 44 |
+
"eval_runtime": 75.0695,
|
| 45 |
+
"eval_samples_per_second": 13.321,
|
| 46 |
+
"eval_steps_per_second": 4.449,
|
| 47 |
+
"step": 2000
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.08,
|
| 51 |
+
"learning_rate": 2.917679212354704e-05,
|
| 52 |
+
"loss": 0.8433,
|
| 53 |
+
"step": 2500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.1,
|
| 57 |
+
"learning_rate": 2.9012150548256447e-05,
|
| 58 |
+
"loss": 0.8052,
|
| 59 |
+
"step": 3000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.12,
|
| 63 |
+
"learning_rate": 2.8847508972965855e-05,
|
| 64 |
+
"loss": 0.8465,
|
| 65 |
+
"step": 3500
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.13,
|
| 69 |
+
"learning_rate": 2.8682867397675263e-05,
|
| 70 |
+
"loss": 0.7535,
|
| 71 |
+
"step": 4000
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.13,
|
| 75 |
+
"eval_exact_match": 13.0,
|
| 76 |
+
"eval_f1": 21.7903,
|
| 77 |
+
"eval_gen_len": 5.483,
|
| 78 |
+
"eval_loss": 0.5836785435676575,
|
| 79 |
+
"eval_rouge1": 21.3055,
|
| 80 |
+
"eval_rouge2": 9.2894,
|
| 81 |
+
"eval_rougeL": 21.2178,
|
| 82 |
+
"eval_rougeLsum": 21.3015,
|
| 83 |
+
"eval_runtime": 76.4558,
|
| 84 |
+
"eval_samples_per_second": 13.079,
|
| 85 |
+
"eval_steps_per_second": 4.369,
|
| 86 |
+
"step": 4000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.15,
|
| 90 |
+
"learning_rate": 2.851822582238467e-05,
|
| 91 |
+
"loss": 0.8144,
|
| 92 |
+
"step": 4500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.16,
|
| 96 |
+
"learning_rate": 2.8353584247094075e-05,
|
| 97 |
+
"loss": 0.7393,
|
| 98 |
+
"step": 5000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.18,
|
| 102 |
+
"learning_rate": 2.8188942671803483e-05,
|
| 103 |
+
"loss": 0.7273,
|
| 104 |
+
"step": 5500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.2,
|
| 108 |
+
"learning_rate": 2.802430109651289e-05,
|
| 109 |
+
"loss": 0.6731,
|
| 110 |
+
"step": 6000
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.2,
|
| 114 |
+
"eval_exact_match": 11.9,
|
| 115 |
+
"eval_f1": 21.595,
|
| 116 |
+
"eval_gen_len": 5.467,
|
| 117 |
+
"eval_loss": 0.5862796306610107,
|
| 118 |
+
"eval_rouge1": 21.739,
|
| 119 |
+
"eval_rouge2": 9.5334,
|
| 120 |
+
"eval_rougeL": 21.7018,
|
| 121 |
+
"eval_rougeLsum": 21.6884,
|
| 122 |
+
"eval_runtime": 74.4712,
|
| 123 |
+
"eval_samples_per_second": 13.428,
|
| 124 |
+
"eval_steps_per_second": 4.485,
|
| 125 |
+
"step": 6000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.21,
|
| 129 |
+
"learning_rate": 2.7859659521222298e-05,
|
| 130 |
+
"loss": 0.6948,
|
| 131 |
+
"step": 6500
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.23,
|
| 135 |
+
"learning_rate": 2.769501794593171e-05,
|
| 136 |
+
"loss": 0.6698,
|
| 137 |
+
"step": 7000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.25,
|
| 141 |
+
"learning_rate": 2.7530376370641117e-05,
|
| 142 |
+
"loss": 0.6466,
|
| 143 |
+
"step": 7500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.26,
|
| 147 |
+
"learning_rate": 2.7365734795350525e-05,
|
| 148 |
+
"loss": 0.6879,
|
| 149 |
+
"step": 8000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.26,
|
| 153 |
+
"eval_exact_match": 6.1,
|
| 154 |
+
"eval_f1": 22.3924,
|
| 155 |
+
"eval_gen_len": 5.95,
|
| 156 |
+
"eval_loss": 0.5380619168281555,
|
| 157 |
+
"eval_rouge1": 25.3106,
|
| 158 |
+
"eval_rouge2": 11.8582,
|
| 159 |
+
"eval_rougeL": 25.2639,
|
| 160 |
+
"eval_rougeLsum": 25.2861,
|
| 161 |
+
"eval_runtime": 76.6994,
|
| 162 |
+
"eval_samples_per_second": 13.038,
|
| 163 |
+
"eval_steps_per_second": 4.355,
|
| 164 |
+
"step": 8000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.28,
|
| 168 |
+
"learning_rate": 2.720109322005993e-05,
|
| 169 |
+
"loss": 0.6504,
|
| 170 |
+
"step": 8500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3,
|
| 174 |
+
"learning_rate": 2.7036451644769337e-05,
|
| 175 |
+
"loss": 0.6834,
|
| 176 |
+
"step": 9000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.31,
|
| 180 |
+
"learning_rate": 2.6871810069478745e-05,
|
| 181 |
+
"loss": 0.6759,
|
| 182 |
+
"step": 9500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.33,
|
| 186 |
+
"learning_rate": 2.6707168494188152e-05,
|
| 187 |
+
"loss": 0.6459,
|
| 188 |
+
"step": 10000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.33,
|
| 192 |
+
"eval_exact_match": 8.1,
|
| 193 |
+
"eval_f1": 22.3008,
|
| 194 |
+
"eval_gen_len": 5.583,
|
| 195 |
+
"eval_loss": 0.536933958530426,
|
| 196 |
+
"eval_rouge1": 25.1248,
|
| 197 |
+
"eval_rouge2": 11.4572,
|
| 198 |
+
"eval_rougeL": 25.13,
|
| 199 |
+
"eval_rougeLsum": 25.1636,
|
| 200 |
+
"eval_runtime": 74.8336,
|
| 201 |
+
"eval_samples_per_second": 13.363,
|
| 202 |
+
"eval_steps_per_second": 4.463,
|
| 203 |
+
"step": 10000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 0.35,
|
| 207 |
+
"learning_rate": 2.654252691889756e-05,
|
| 208 |
+
"loss": 0.6555,
|
| 209 |
+
"step": 10500
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.36,
|
| 213 |
+
"learning_rate": 2.6377885343606968e-05,
|
| 214 |
+
"loss": 0.622,
|
| 215 |
+
"step": 11000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.38,
|
| 219 |
+
"learning_rate": 2.6213243768316376e-05,
|
| 220 |
+
"loss": 0.6434,
|
| 221 |
+
"step": 11500
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.4,
|
| 225 |
+
"learning_rate": 2.6048602193025783e-05,
|
| 226 |
+
"loss": 0.5865,
|
| 227 |
+
"step": 12000
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.4,
|
| 231 |
+
"eval_exact_match": 7.8,
|
| 232 |
+
"eval_f1": 22.5664,
|
| 233 |
+
"eval_gen_len": 5.502,
|
| 234 |
+
"eval_loss": 0.5340938568115234,
|
| 235 |
+
"eval_rouge1": 25.6756,
|
| 236 |
+
"eval_rouge2": 11.0984,
|
| 237 |
+
"eval_rougeL": 25.6163,
|
| 238 |
+
"eval_rougeLsum": 25.6348,
|
| 239 |
+
"eval_runtime": 74.1837,
|
| 240 |
+
"eval_samples_per_second": 13.48,
|
| 241 |
+
"eval_steps_per_second": 4.502,
|
| 242 |
+
"step": 12000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.41,
|
| 246 |
+
"learning_rate": 2.588396061773519e-05,
|
| 247 |
+
"loss": 0.573,
|
| 248 |
+
"step": 12500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.43,
|
| 252 |
+
"learning_rate": 2.57193190424446e-05,
|
| 253 |
+
"loss": 0.6825,
|
| 254 |
+
"step": 13000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.44,
|
| 258 |
+
"learning_rate": 2.5554677467154007e-05,
|
| 259 |
+
"loss": 0.5871,
|
| 260 |
+
"step": 13500
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.46,
|
| 264 |
+
"learning_rate": 2.5390035891863414e-05,
|
| 265 |
+
"loss": 0.6442,
|
| 266 |
+
"step": 14000
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.46,
|
| 270 |
+
"eval_exact_match": 8.0,
|
| 271 |
+
"eval_f1": 22.9873,
|
| 272 |
+
"eval_gen_len": 5.694,
|
| 273 |
+
"eval_loss": 0.5320589542388916,
|
| 274 |
+
"eval_rouge1": 25.6004,
|
| 275 |
+
"eval_rouge2": 11.31,
|
| 276 |
+
"eval_rougeL": 25.6195,
|
| 277 |
+
"eval_rougeLsum": 25.6027,
|
| 278 |
+
"eval_runtime": 75.839,
|
| 279 |
+
"eval_samples_per_second": 13.186,
|
| 280 |
+
"eval_steps_per_second": 4.404,
|
| 281 |
+
"step": 14000
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.48,
|
| 285 |
+
"learning_rate": 2.5225394316572822e-05,
|
| 286 |
+
"loss": 0.6252,
|
| 287 |
+
"step": 14500
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.49,
|
| 291 |
+
"learning_rate": 2.5060752741282227e-05,
|
| 292 |
+
"loss": 0.613,
|
| 293 |
+
"step": 15000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 0.51,
|
| 297 |
+
"learning_rate": 2.4896111165991638e-05,
|
| 298 |
+
"loss": 0.5827,
|
| 299 |
+
"step": 15500
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.53,
|
| 303 |
+
"learning_rate": 2.4731469590701045e-05,
|
| 304 |
+
"loss": 0.5819,
|
| 305 |
+
"step": 16000
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.53,
|
| 309 |
+
"eval_exact_match": 6.6,
|
| 310 |
+
"eval_f1": 23.22,
|
| 311 |
+
"eval_gen_len": 5.789,
|
| 312 |
+
"eval_loss": 0.5230505466461182,
|
| 313 |
+
"eval_rouge1": 25.3937,
|
| 314 |
+
"eval_rouge2": 11.1344,
|
| 315 |
+
"eval_rougeL": 25.3304,
|
| 316 |
+
"eval_rougeLsum": 25.3461,
|
| 317 |
+
"eval_runtime": 75.9506,
|
| 318 |
+
"eval_samples_per_second": 13.166,
|
| 319 |
+
"eval_steps_per_second": 4.398,
|
| 320 |
+
"step": 16000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.54,
|
| 324 |
+
"learning_rate": 2.4566828015410453e-05,
|
| 325 |
+
"loss": 0.6175,
|
| 326 |
+
"step": 16500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.56,
|
| 330 |
+
"learning_rate": 2.440218644011986e-05,
|
| 331 |
+
"loss": 0.6132,
|
| 332 |
+
"step": 17000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.58,
|
| 336 |
+
"learning_rate": 2.423754486482927e-05,
|
| 337 |
+
"loss": 0.6048,
|
| 338 |
+
"step": 17500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.59,
|
| 342 |
+
"learning_rate": 2.4072903289538673e-05,
|
| 343 |
+
"loss": 0.5908,
|
| 344 |
+
"step": 18000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.59,
|
| 348 |
+
"eval_exact_match": 7.3,
|
| 349 |
+
"eval_f1": 23.0102,
|
| 350 |
+
"eval_gen_len": 5.769,
|
| 351 |
+
"eval_loss": 0.5285995006561279,
|
| 352 |
+
"eval_rouge1": 25.898,
|
| 353 |
+
"eval_rouge2": 11.687,
|
| 354 |
+
"eval_rougeL": 25.8492,
|
| 355 |
+
"eval_rougeLsum": 25.8886,
|
| 356 |
+
"eval_runtime": 76.3215,
|
| 357 |
+
"eval_samples_per_second": 13.102,
|
| 358 |
+
"eval_steps_per_second": 4.376,
|
| 359 |
+
"step": 18000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.61,
|
| 363 |
+
"learning_rate": 2.390826171424808e-05,
|
| 364 |
+
"loss": 0.5972,
|
| 365 |
+
"step": 18500
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.63,
|
| 369 |
+
"learning_rate": 2.374362013895749e-05,
|
| 370 |
+
"loss": 0.5995,
|
| 371 |
+
"step": 19000
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 0.64,
|
| 375 |
+
"learning_rate": 2.3578978563666896e-05,
|
| 376 |
+
"loss": 0.6133,
|
| 377 |
+
"step": 19500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.66,
|
| 381 |
+
"learning_rate": 2.3414336988376307e-05,
|
| 382 |
+
"loss": 0.5814,
|
| 383 |
+
"step": 20000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.66,
|
| 387 |
+
"eval_exact_match": 6.8,
|
| 388 |
+
"eval_f1": 23.0215,
|
| 389 |
+
"eval_gen_len": 5.956,
|
| 390 |
+
"eval_loss": 0.5266488790512085,
|
| 391 |
+
"eval_rouge1": 25.7545,
|
| 392 |
+
"eval_rouge2": 11.6372,
|
| 393 |
+
"eval_rougeL": 25.7025,
|
| 394 |
+
"eval_rougeLsum": 25.6994,
|
| 395 |
+
"eval_runtime": 77.5697,
|
| 396 |
+
"eval_samples_per_second": 12.892,
|
| 397 |
+
"eval_steps_per_second": 4.306,
|
| 398 |
+
"step": 20000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.68,
|
| 402 |
+
"learning_rate": 2.3249695413085715e-05,
|
| 403 |
+
"loss": 0.5501,
|
| 404 |
+
"step": 20500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.69,
|
| 408 |
+
"learning_rate": 2.3085053837795123e-05,
|
| 409 |
+
"loss": 0.5771,
|
| 410 |
+
"step": 21000
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.71,
|
| 414 |
+
"learning_rate": 2.2920412262504527e-05,
|
| 415 |
+
"loss": 0.5856,
|
| 416 |
+
"step": 21500
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.72,
|
| 420 |
+
"learning_rate": 2.2755770687213935e-05,
|
| 421 |
+
"loss": 0.5969,
|
| 422 |
+
"step": 22000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.72,
|
| 426 |
+
"eval_exact_match": 6.6,
|
| 427 |
+
"eval_f1": 22.3945,
|
| 428 |
+
"eval_gen_len": 5.784,
|
| 429 |
+
"eval_loss": 0.5150488018989563,
|
| 430 |
+
"eval_rouge1": 26.376,
|
| 431 |
+
"eval_rouge2": 11.4531,
|
| 432 |
+
"eval_rougeL": 26.3401,
|
| 433 |
+
"eval_rougeLsum": 26.3444,
|
| 434 |
+
"eval_runtime": 76.9076,
|
| 435 |
+
"eval_samples_per_second": 13.003,
|
| 436 |
+
"eval_steps_per_second": 4.343,
|
| 437 |
+
"step": 22000
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.74,
|
| 441 |
+
"learning_rate": 2.2591129111923343e-05,
|
| 442 |
+
"loss": 0.6051,
|
| 443 |
+
"step": 22500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.76,
|
| 447 |
+
"learning_rate": 2.242648753663275e-05,
|
| 448 |
+
"loss": 0.6252,
|
| 449 |
+
"step": 23000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.77,
|
| 453 |
+
"learning_rate": 2.226184596134216e-05,
|
| 454 |
+
"loss": 0.6002,
|
| 455 |
+
"step": 23500
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 0.79,
|
| 459 |
+
"learning_rate": 2.2097204386051566e-05,
|
| 460 |
+
"loss": 0.6071,
|
| 461 |
+
"step": 24000
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.79,
|
| 465 |
+
"eval_exact_match": 7.5,
|
| 466 |
+
"eval_f1": 22.429,
|
| 467 |
+
"eval_gen_len": 5.577,
|
| 468 |
+
"eval_loss": 0.5234242081642151,
|
| 469 |
+
"eval_rouge1": 26.0675,
|
| 470 |
+
"eval_rouge2": 11.2086,
|
| 471 |
+
"eval_rougeL": 26.062,
|
| 472 |
+
"eval_rougeLsum": 26.0623,
|
| 473 |
+
"eval_runtime": 74.7901,
|
| 474 |
+
"eval_samples_per_second": 13.371,
|
| 475 |
+
"eval_steps_per_second": 4.466,
|
| 476 |
+
"step": 24000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.81,
|
| 480 |
+
"learning_rate": 2.1932562810760974e-05,
|
| 481 |
+
"loss": 0.5639,
|
| 482 |
+
"step": 24500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.82,
|
| 486 |
+
"learning_rate": 2.176792123547038e-05,
|
| 487 |
+
"loss": 0.5438,
|
| 488 |
+
"step": 25000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.84,
|
| 492 |
+
"learning_rate": 2.160327966017979e-05,
|
| 493 |
+
"loss": 0.5473,
|
| 494 |
+
"step": 25500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.86,
|
| 498 |
+
"learning_rate": 2.1438638084889197e-05,
|
| 499 |
+
"loss": 0.5743,
|
| 500 |
+
"step": 26000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.86,
|
| 504 |
+
"eval_exact_match": 6.1,
|
| 505 |
+
"eval_f1": 22.4076,
|
| 506 |
+
"eval_gen_len": 5.714,
|
| 507 |
+
"eval_loss": 0.5406718850135803,
|
| 508 |
+
"eval_rouge1": 27.3834,
|
| 509 |
+
"eval_rouge2": 11.9993,
|
| 510 |
+
"eval_rougeL": 27.3736,
|
| 511 |
+
"eval_rougeLsum": 27.3494,
|
| 512 |
+
"eval_runtime": 75.5929,
|
| 513 |
+
"eval_samples_per_second": 13.229,
|
| 514 |
+
"eval_steps_per_second": 4.418,
|
| 515 |
+
"step": 26000
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.87,
|
| 519 |
+
"learning_rate": 2.1273996509598605e-05,
|
| 520 |
+
"loss": 0.6123,
|
| 521 |
+
"step": 26500
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.89,
|
| 525 |
+
"learning_rate": 2.1109354934308013e-05,
|
| 526 |
+
"loss": 0.5416,
|
| 527 |
+
"step": 27000
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.91,
|
| 531 |
+
"learning_rate": 2.094471335901742e-05,
|
| 532 |
+
"loss": 0.5794,
|
| 533 |
+
"step": 27500
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 0.92,
|
| 537 |
+
"learning_rate": 2.0780071783726825e-05,
|
| 538 |
+
"loss": 0.5513,
|
| 539 |
+
"step": 28000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 0.92,
|
| 543 |
+
"eval_exact_match": 5.6,
|
| 544 |
+
"eval_f1": 23.3636,
|
| 545 |
+
"eval_gen_len": 5.841,
|
| 546 |
+
"eval_loss": 0.516300618648529,
|
| 547 |
+
"eval_rouge1": 27.5598,
|
| 548 |
+
"eval_rouge2": 12.4259,
|
| 549 |
+
"eval_rougeL": 27.494,
|
| 550 |
+
"eval_rougeLsum": 27.5139,
|
| 551 |
+
"eval_runtime": 75.7363,
|
| 552 |
+
"eval_samples_per_second": 13.204,
|
| 553 |
+
"eval_steps_per_second": 4.41,
|
| 554 |
+
"step": 28000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.94,
|
| 558 |
+
"learning_rate": 2.0615430208436232e-05,
|
| 559 |
+
"loss": 0.5198,
|
| 560 |
+
"step": 28500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.95,
|
| 564 |
+
"learning_rate": 2.0450788633145644e-05,
|
| 565 |
+
"loss": 0.5419,
|
| 566 |
+
"step": 29000
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.97,
|
| 570 |
+
"learning_rate": 2.028614705785505e-05,
|
| 571 |
+
"loss": 0.594,
|
| 572 |
+
"step": 29500
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.99,
|
| 576 |
+
"learning_rate": 2.012150548256446e-05,
|
| 577 |
+
"loss": 0.5916,
|
| 578 |
+
"step": 30000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.99,
|
| 582 |
+
"eval_exact_match": 6.5,
|
| 583 |
+
"eval_f1": 23.2412,
|
| 584 |
+
"eval_gen_len": 5.805,
|
| 585 |
+
"eval_loss": 0.524455726146698,
|
| 586 |
+
"eval_rouge1": 26.6244,
|
| 587 |
+
"eval_rouge2": 11.8352,
|
| 588 |
+
"eval_rougeL": 26.5432,
|
| 589 |
+
"eval_rougeLsum": 26.5964,
|
| 590 |
+
"eval_runtime": 76.4476,
|
| 591 |
+
"eval_samples_per_second": 13.081,
|
| 592 |
+
"eval_steps_per_second": 4.369,
|
| 593 |
+
"step": 30000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 1.0,
|
| 597 |
+
"learning_rate": 1.9956863907273867e-05,
|
| 598 |
+
"loss": 0.5885,
|
| 599 |
+
"step": 30500
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 1.02,
|
| 603 |
+
"learning_rate": 1.979222233198327e-05,
|
| 604 |
+
"loss": 0.5536,
|
| 605 |
+
"step": 31000
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.04,
|
| 609 |
+
"learning_rate": 1.962758075669268e-05,
|
| 610 |
+
"loss": 0.5119,
|
| 611 |
+
"step": 31500
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.05,
|
| 615 |
+
"learning_rate": 1.9462939181402087e-05,
|
| 616 |
+
"loss": 0.5046,
|
| 617 |
+
"step": 32000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 1.05,
|
| 621 |
+
"eval_exact_match": 8.6,
|
| 622 |
+
"eval_f1": 22.4289,
|
| 623 |
+
"eval_gen_len": 5.578,
|
| 624 |
+
"eval_loss": 0.5430198907852173,
|
| 625 |
+
"eval_rouge1": 26.57,
|
| 626 |
+
"eval_rouge2": 11.7677,
|
| 627 |
+
"eval_rougeL": 26.6056,
|
| 628 |
+
"eval_rougeLsum": 26.5575,
|
| 629 |
+
"eval_runtime": 74.902,
|
| 630 |
+
"eval_samples_per_second": 13.351,
|
| 631 |
+
"eval_steps_per_second": 4.459,
|
| 632 |
+
"step": 32000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.07,
|
| 636 |
+
"learning_rate": 1.9298297606111494e-05,
|
| 637 |
+
"loss": 0.5077,
|
| 638 |
+
"step": 32500
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 1.09,
|
| 642 |
+
"learning_rate": 1.9133656030820902e-05,
|
| 643 |
+
"loss": 0.5206,
|
| 644 |
+
"step": 33000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 1.1,
|
| 648 |
+
"learning_rate": 1.8969014455530313e-05,
|
| 649 |
+
"loss": 0.5079,
|
| 650 |
+
"step": 33500
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 1.12,
|
| 654 |
+
"learning_rate": 1.880437288023972e-05,
|
| 655 |
+
"loss": 0.5349,
|
| 656 |
+
"step": 34000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 1.12,
|
| 660 |
+
"eval_exact_match": 5.4,
|
| 661 |
+
"eval_f1": 23.2951,
|
| 662 |
+
"eval_gen_len": 5.744,
|
| 663 |
+
"eval_loss": 0.517284631729126,
|
| 664 |
+
"eval_rouge1": 28.0046,
|
| 665 |
+
"eval_rouge2": 12.2802,
|
| 666 |
+
"eval_rougeL": 28.0025,
|
| 667 |
+
"eval_rougeLsum": 27.9706,
|
| 668 |
+
"eval_runtime": 75.9315,
|
| 669 |
+
"eval_samples_per_second": 13.17,
|
| 670 |
+
"eval_steps_per_second": 4.399,
|
| 671 |
+
"step": 34000
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 1.14,
|
| 675 |
+
"learning_rate": 1.8639731304949125e-05,
|
| 676 |
+
"loss": 0.4871,
|
| 677 |
+
"step": 34500
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 1.15,
|
| 681 |
+
"learning_rate": 1.8475089729658533e-05,
|
| 682 |
+
"loss": 0.541,
|
| 683 |
+
"step": 35000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 1.17,
|
| 687 |
+
"learning_rate": 1.831044815436794e-05,
|
| 688 |
+
"loss": 0.5659,
|
| 689 |
+
"step": 35500
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.19,
|
| 693 |
+
"learning_rate": 1.814580657907735e-05,
|
| 694 |
+
"loss": 0.5231,
|
| 695 |
+
"step": 36000
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.19,
|
| 699 |
+
"eval_exact_match": 6.1,
|
| 700 |
+
"eval_f1": 23.2074,
|
| 701 |
+
"eval_gen_len": 5.814,
|
| 702 |
+
"eval_loss": 0.5330841541290283,
|
| 703 |
+
"eval_rouge1": 27.4091,
|
| 704 |
+
"eval_rouge2": 12.3926,
|
| 705 |
+
"eval_rougeL": 27.3466,
|
| 706 |
+
"eval_rougeLsum": 27.3708,
|
| 707 |
+
"eval_runtime": 77.6684,
|
| 708 |
+
"eval_samples_per_second": 12.875,
|
| 709 |
+
"eval_steps_per_second": 4.3,
|
| 710 |
+
"step": 36000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.2,
|
| 714 |
+
"learning_rate": 1.7981165003786756e-05,
|
| 715 |
+
"loss": 0.5209,
|
| 716 |
+
"step": 36500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.22,
|
| 720 |
+
"learning_rate": 1.7816523428496164e-05,
|
| 721 |
+
"loss": 0.5561,
|
| 722 |
+
"step": 37000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 1.23,
|
| 726 |
+
"learning_rate": 1.7651881853205572e-05,
|
| 727 |
+
"loss": 0.5269,
|
| 728 |
+
"step": 37500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 1.25,
|
| 732 |
+
"learning_rate": 1.748724027791498e-05,
|
| 733 |
+
"loss": 0.5144,
|
| 734 |
+
"step": 38000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.25,
|
| 738 |
+
"eval_exact_match": 6.6,
|
| 739 |
+
"eval_f1": 22.9561,
|
| 740 |
+
"eval_gen_len": 5.764,
|
| 741 |
+
"eval_loss": 0.533650815486908,
|
| 742 |
+
"eval_rouge1": 27.7074,
|
| 743 |
+
"eval_rouge2": 12.5122,
|
| 744 |
+
"eval_rougeL": 27.6729,
|
| 745 |
+
"eval_rougeLsum": 27.6909,
|
| 746 |
+
"eval_runtime": 76.8102,
|
| 747 |
+
"eval_samples_per_second": 13.019,
|
| 748 |
+
"eval_steps_per_second": 4.348,
|
| 749 |
+
"step": 38000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 1.27,
|
| 753 |
+
"learning_rate": 1.7322598702624387e-05,
|
| 754 |
+
"loss": 0.4877,
|
| 755 |
+
"step": 38500
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 1.28,
|
| 759 |
+
"learning_rate": 1.7157957127333795e-05,
|
| 760 |
+
"loss": 0.5182,
|
| 761 |
+
"step": 39000
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 1.3,
|
| 765 |
+
"learning_rate": 1.6993315552043203e-05,
|
| 766 |
+
"loss": 0.4982,
|
| 767 |
+
"step": 39500
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.32,
|
| 771 |
+
"learning_rate": 1.682867397675261e-05,
|
| 772 |
+
"loss": 0.5576,
|
| 773 |
+
"step": 40000
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.32,
|
| 777 |
+
"eval_exact_match": 7.0,
|
| 778 |
+
"eval_f1": 22.646,
|
| 779 |
+
"eval_gen_len": 5.801,
|
| 780 |
+
"eval_loss": 0.5152503848075867,
|
| 781 |
+
"eval_rouge1": 27.8201,
|
| 782 |
+
"eval_rouge2": 12.335,
|
| 783 |
+
"eval_rougeL": 27.806,
|
| 784 |
+
"eval_rougeLsum": 27.7887,
|
| 785 |
+
"eval_runtime": 77.4695,
|
| 786 |
+
"eval_samples_per_second": 12.908,
|
| 787 |
+
"eval_steps_per_second": 4.311,
|
| 788 |
+
"step": 40000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 1.33,
|
| 792 |
+
"learning_rate": 1.666403240146202e-05,
|
| 793 |
+
"loss": 0.5032,
|
| 794 |
+
"step": 40500
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.35,
|
| 798 |
+
"learning_rate": 1.6499390826171423e-05,
|
| 799 |
+
"loss": 0.5397,
|
| 800 |
+
"step": 41000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.37,
|
| 804 |
+
"learning_rate": 1.633474925088083e-05,
|
| 805 |
+
"loss": 0.4944,
|
| 806 |
+
"step": 41500
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 1.38,
|
| 810 |
+
"learning_rate": 1.617010767559024e-05,
|
| 811 |
+
"loss": 0.5184,
|
| 812 |
+
"step": 42000
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 1.38,
|
| 816 |
+
"eval_exact_match": 4.8,
|
| 817 |
+
"eval_f1": 23.3492,
|
| 818 |
+
"eval_gen_len": 5.988,
|
| 819 |
+
"eval_loss": 0.5299638509750366,
|
| 820 |
+
"eval_rouge1": 28.1638,
|
| 821 |
+
"eval_rouge2": 12.5269,
|
| 822 |
+
"eval_rougeL": 28.1245,
|
| 823 |
+
"eval_rougeLsum": 28.1199,
|
| 824 |
+
"eval_runtime": 77.8204,
|
| 825 |
+
"eval_samples_per_second": 12.85,
|
| 826 |
+
"eval_steps_per_second": 4.292,
|
| 827 |
+
"step": 42000
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 1.4,
|
| 831 |
+
"learning_rate": 1.600546610029965e-05,
|
| 832 |
+
"loss": 0.488,
|
| 833 |
+
"step": 42500
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 1.42,
|
| 837 |
+
"learning_rate": 1.5840824525009057e-05,
|
| 838 |
+
"loss": 0.5176,
|
| 839 |
+
"step": 43000
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 1.43,
|
| 843 |
+
"learning_rate": 1.5676182949718465e-05,
|
| 844 |
+
"loss": 0.4941,
|
| 845 |
+
"step": 43500
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 1.45,
|
| 849 |
+
"learning_rate": 1.5511541374427873e-05,
|
| 850 |
+
"loss": 0.5321,
|
| 851 |
+
"step": 44000
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 1.45,
|
| 855 |
+
"eval_exact_match": 5.2,
|
| 856 |
+
"eval_f1": 23.1651,
|
| 857 |
+
"eval_gen_len": 5.997,
|
| 858 |
+
"eval_loss": 0.5239735245704651,
|
| 859 |
+
"eval_rouge1": 27.9328,
|
| 860 |
+
"eval_rouge2": 12.5264,
|
| 861 |
+
"eval_rougeL": 27.9238,
|
| 862 |
+
"eval_rougeLsum": 27.9206,
|
| 863 |
+
"eval_runtime": 78.5078,
|
| 864 |
+
"eval_samples_per_second": 12.738,
|
| 865 |
+
"eval_steps_per_second": 4.254,
|
| 866 |
+
"step": 44000
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 1.47,
|
| 870 |
+
"learning_rate": 1.5346899799137277e-05,
|
| 871 |
+
"loss": 0.5,
|
| 872 |
+
"step": 44500
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.48,
|
| 876 |
+
"learning_rate": 1.5182258223846685e-05,
|
| 877 |
+
"loss": 0.5279,
|
| 878 |
+
"step": 45000
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.5,
|
| 882 |
+
"learning_rate": 1.5017616648556093e-05,
|
| 883 |
+
"loss": 0.5223,
|
| 884 |
+
"step": 45500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.51,
|
| 888 |
+
"learning_rate": 1.4852975073265502e-05,
|
| 889 |
+
"loss": 0.478,
|
| 890 |
+
"step": 46000
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 1.51,
|
| 894 |
+
"eval_exact_match": 6.4,
|
| 895 |
+
"eval_f1": 22.7142,
|
| 896 |
+
"eval_gen_len": 5.824,
|
| 897 |
+
"eval_loss": 0.5349439382553101,
|
| 898 |
+
"eval_rouge1": 27.9589,
|
| 899 |
+
"eval_rouge2": 12.5606,
|
| 900 |
+
"eval_rougeL": 27.902,
|
| 901 |
+
"eval_rougeLsum": 27.8978,
|
| 902 |
+
"eval_runtime": 76.8435,
|
| 903 |
+
"eval_samples_per_second": 13.013,
|
| 904 |
+
"eval_steps_per_second": 4.346,
|
| 905 |
+
"step": 46000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.53,
|
| 909 |
+
"learning_rate": 1.4688333497974908e-05,
|
| 910 |
+
"loss": 0.482,
|
| 911 |
+
"step": 46500
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 1.55,
|
| 915 |
+
"learning_rate": 1.4523691922684316e-05,
|
| 916 |
+
"loss": 0.5004,
|
| 917 |
+
"step": 47000
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 1.56,
|
| 921 |
+
"learning_rate": 1.4359050347393724e-05,
|
| 922 |
+
"loss": 0.5482,
|
| 923 |
+
"step": 47500
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 1.58,
|
| 927 |
+
"learning_rate": 1.4194408772103133e-05,
|
| 928 |
+
"loss": 0.497,
|
| 929 |
+
"step": 48000
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 1.58,
|
| 933 |
+
"eval_exact_match": 5.5,
|
| 934 |
+
"eval_f1": 22.6861,
|
| 935 |
+
"eval_gen_len": 5.864,
|
| 936 |
+
"eval_loss": 0.5399107933044434,
|
| 937 |
+
"eval_rouge1": 28.2327,
|
| 938 |
+
"eval_rouge2": 12.5728,
|
| 939 |
+
"eval_rougeL": 28.1968,
|
| 940 |
+
"eval_rougeLsum": 28.21,
|
| 941 |
+
"eval_runtime": 77.9,
|
| 942 |
+
"eval_samples_per_second": 12.837,
|
| 943 |
+
"eval_steps_per_second": 4.288,
|
| 944 |
+
"step": 48000
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 1.6,
|
| 948 |
+
"learning_rate": 1.4029767196812539e-05,
|
| 949 |
+
"loss": 0.4476,
|
| 950 |
+
"step": 48500
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 1.61,
|
| 954 |
+
"learning_rate": 1.3865125621521947e-05,
|
| 955 |
+
"loss": 0.5754,
|
| 956 |
+
"step": 49000
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 1.63,
|
| 960 |
+
"learning_rate": 1.3700484046231355e-05,
|
| 961 |
+
"loss": 0.5296,
|
| 962 |
+
"step": 49500
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.65,
|
| 966 |
+
"learning_rate": 1.3535842470940762e-05,
|
| 967 |
+
"loss": 0.5475,
|
| 968 |
+
"step": 50000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.65,
|
| 972 |
+
"eval_exact_match": 8.0,
|
| 973 |
+
"eval_f1": 22.2882,
|
| 974 |
+
"eval_gen_len": 5.651,
|
| 975 |
+
"eval_loss": 0.5568350553512573,
|
| 976 |
+
"eval_rouge1": 27.6,
|
| 977 |
+
"eval_rouge2": 12.259,
|
| 978 |
+
"eval_rougeL": 27.6081,
|
| 979 |
+
"eval_rougeLsum": 27.561,
|
| 980 |
+
"eval_runtime": 76.4106,
|
| 981 |
+
"eval_samples_per_second": 13.087,
|
| 982 |
+
"eval_steps_per_second": 4.371,
|
| 983 |
+
"step": 50000
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.66,
|
| 987 |
+
"learning_rate": 1.337120089565017e-05,
|
| 988 |
+
"loss": 0.492,
|
| 989 |
+
"step": 50500
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.68,
|
| 993 |
+
"learning_rate": 1.3206559320359578e-05,
|
| 994 |
+
"loss": 0.5221,
|
| 995 |
+
"step": 51000
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 1.7,
|
| 999 |
+
"learning_rate": 1.3041917745068984e-05,
|
| 1000 |
+
"loss": 0.5491,
|
| 1001 |
+
"step": 51500
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 1.71,
|
| 1005 |
+
"learning_rate": 1.2877276169778392e-05,
|
| 1006 |
+
"loss": 0.4643,
|
| 1007 |
+
"step": 52000
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 1.71,
|
| 1011 |
+
"eval_exact_match": 8.9,
|
| 1012 |
+
"eval_f1": 21.3228,
|
| 1013 |
+
"eval_gen_len": 5.566,
|
| 1014 |
+
"eval_loss": 0.5837388038635254,
|
| 1015 |
+
"eval_rouge1": 27.0269,
|
| 1016 |
+
"eval_rouge2": 11.9442,
|
| 1017 |
+
"eval_rougeL": 27.0247,
|
| 1018 |
+
"eval_rougeLsum": 26.9566,
|
| 1019 |
+
"eval_runtime": 76.2944,
|
| 1020 |
+
"eval_samples_per_second": 13.107,
|
| 1021 |
+
"eval_steps_per_second": 4.378,
|
| 1022 |
+
"step": 52000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 1.73,
|
| 1026 |
+
"learning_rate": 1.2712634594487801e-05,
|
| 1027 |
+
"loss": 0.5079,
|
| 1028 |
+
"step": 52500
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 1.75,
|
| 1032 |
+
"learning_rate": 1.2547993019197209e-05,
|
| 1033 |
+
"loss": 0.4905,
|
| 1034 |
+
"step": 53000
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 1.76,
|
| 1038 |
+
"learning_rate": 1.2383351443906615e-05,
|
| 1039 |
+
"loss": 0.4939,
|
| 1040 |
+
"step": 53500
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 1.78,
|
| 1044 |
+
"learning_rate": 1.2218709868616023e-05,
|
| 1045 |
+
"loss": 0.5333,
|
| 1046 |
+
"step": 54000
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.78,
|
| 1050 |
+
"eval_exact_match": 5.5,
|
| 1051 |
+
"eval_f1": 23.1489,
|
| 1052 |
+
"eval_gen_len": 5.764,
|
| 1053 |
+
"eval_loss": 0.5246403813362122,
|
| 1054 |
+
"eval_rouge1": 27.9369,
|
| 1055 |
+
"eval_rouge2": 12.2044,
|
| 1056 |
+
"eval_rougeL": 27.8886,
|
| 1057 |
+
"eval_rougeLsum": 27.9162,
|
| 1058 |
+
"eval_runtime": 76.3987,
|
| 1059 |
+
"eval_samples_per_second": 13.089,
|
| 1060 |
+
"eval_steps_per_second": 4.372,
|
| 1061 |
+
"step": 54000
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 1.79,
|
| 1065 |
+
"learning_rate": 1.2054068293325432e-05,
|
| 1066 |
+
"loss": 0.5067,
|
| 1067 |
+
"step": 54500
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.81,
|
| 1071 |
+
"learning_rate": 1.1889426718034838e-05,
|
| 1072 |
+
"loss": 0.4915,
|
| 1073 |
+
"step": 55000
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.83,
|
| 1077 |
+
"learning_rate": 1.1724785142744246e-05,
|
| 1078 |
+
"loss": 0.4855,
|
| 1079 |
+
"step": 55500
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 1.84,
|
| 1083 |
+
"learning_rate": 1.1560143567453654e-05,
|
| 1084 |
+
"loss": 0.5356,
|
| 1085 |
+
"step": 56000
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 1.84,
|
| 1089 |
+
"eval_exact_match": 7.5,
|
| 1090 |
+
"eval_f1": 21.9492,
|
| 1091 |
+
"eval_gen_len": 5.628,
|
| 1092 |
+
"eval_loss": 0.559809684753418,
|
| 1093 |
+
"eval_rouge1": 27.8669,
|
| 1094 |
+
"eval_rouge2": 12.3269,
|
| 1095 |
+
"eval_rougeL": 27.8231,
|
| 1096 |
+
"eval_rougeLsum": 27.8187,
|
| 1097 |
+
"eval_runtime": 75.9003,
|
| 1098 |
+
"eval_samples_per_second": 13.175,
|
| 1099 |
+
"eval_steps_per_second": 4.401,
|
| 1100 |
+
"step": 56000
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 1.86,
|
| 1104 |
+
"learning_rate": 1.1395501992163061e-05,
|
| 1105 |
+
"loss": 0.5376,
|
| 1106 |
+
"step": 56500
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 1.88,
|
| 1110 |
+
"learning_rate": 1.123086041687247e-05,
|
| 1111 |
+
"loss": 0.5152,
|
| 1112 |
+
"step": 57000
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 1.89,
|
| 1116 |
+
"learning_rate": 1.1066218841581877e-05,
|
| 1117 |
+
"loss": 0.5355,
|
| 1118 |
+
"step": 57500
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 1.91,
|
| 1122 |
+
"learning_rate": 1.0901577266291283e-05,
|
| 1123 |
+
"loss": 0.5154,
|
| 1124 |
+
"step": 58000
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 1.91,
|
| 1128 |
+
"eval_exact_match": 5.5,
|
| 1129 |
+
"eval_f1": 22.8791,
|
| 1130 |
+
"eval_gen_len": 5.859,
|
| 1131 |
+
"eval_loss": 0.5113182663917542,
|
| 1132 |
+
"eval_rouge1": 28.1362,
|
| 1133 |
+
"eval_rouge2": 12.2055,
|
| 1134 |
+
"eval_rougeL": 28.0813,
|
| 1135 |
+
"eval_rougeLsum": 28.0901,
|
| 1136 |
+
"eval_runtime": 78.3738,
|
| 1137 |
+
"eval_samples_per_second": 12.759,
|
| 1138 |
+
"eval_steps_per_second": 4.262,
|
| 1139 |
+
"step": 58000
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 1.93,
|
| 1143 |
+
"learning_rate": 1.073693569100069e-05,
|
| 1144 |
+
"loss": 0.5391,
|
| 1145 |
+
"step": 58500
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 1.94,
|
| 1149 |
+
"learning_rate": 1.05722941157101e-05,
|
| 1150 |
+
"loss": 0.4996,
|
| 1151 |
+
"step": 59000
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.96,
|
| 1155 |
+
"learning_rate": 1.0407652540419508e-05,
|
| 1156 |
+
"loss": 0.5221,
|
| 1157 |
+
"step": 59500
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 1.98,
|
| 1161 |
+
"learning_rate": 1.0243010965128914e-05,
|
| 1162 |
+
"loss": 0.4974,
|
| 1163 |
+
"step": 60000
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 1.98,
|
| 1167 |
+
"eval_exact_match": 7.7,
|
| 1168 |
+
"eval_f1": 22.1627,
|
| 1169 |
+
"eval_gen_len": 5.622,
|
| 1170 |
+
"eval_loss": 0.529927670955658,
|
| 1171 |
+
"eval_rouge1": 27.4181,
|
| 1172 |
+
"eval_rouge2": 11.8982,
|
| 1173 |
+
"eval_rougeL": 27.3595,
|
| 1174 |
+
"eval_rougeLsum": 27.362,
|
| 1175 |
+
"eval_runtime": 76.1366,
|
| 1176 |
+
"eval_samples_per_second": 13.134,
|
| 1177 |
+
"eval_steps_per_second": 4.387,
|
| 1178 |
+
"step": 60000
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 1.99,
|
| 1182 |
+
"learning_rate": 1.0078369389838322e-05,
|
| 1183 |
+
"loss": 0.4769,
|
| 1184 |
+
"step": 60500
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 2.01,
|
| 1188 |
+
"learning_rate": 9.913727814547731e-06,
|
| 1189 |
+
"loss": 0.5019,
|
| 1190 |
+
"step": 61000
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 2.03,
|
| 1194 |
+
"learning_rate": 9.749086239257137e-06,
|
| 1195 |
+
"loss": 0.4819,
|
| 1196 |
+
"step": 61500
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 2.04,
|
| 1200 |
+
"learning_rate": 9.584444663966545e-06,
|
| 1201 |
+
"loss": 0.4368,
|
| 1202 |
+
"step": 62000
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 2.04,
|
| 1206 |
+
"eval_exact_match": 4.9,
|
| 1207 |
+
"eval_f1": 23.2707,
|
| 1208 |
+
"eval_gen_len": 5.85,
|
| 1209 |
+
"eval_loss": 0.533755362033844,
|
| 1210 |
+
"eval_rouge1": 28.3591,
|
| 1211 |
+
"eval_rouge2": 12.5844,
|
| 1212 |
+
"eval_rougeL": 28.2897,
|
| 1213 |
+
"eval_rougeLsum": 28.3187,
|
| 1214 |
+
"eval_runtime": 77.3013,
|
| 1215 |
+
"eval_samples_per_second": 12.936,
|
| 1216 |
+
"eval_steps_per_second": 4.321,
|
| 1217 |
+
"step": 62000
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 2.06,
|
| 1221 |
+
"learning_rate": 9.419803088675953e-06,
|
| 1222 |
+
"loss": 0.5444,
|
| 1223 |
+
"step": 62500
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 2.07,
|
| 1227 |
+
"learning_rate": 9.255161513385359e-06,
|
| 1228 |
+
"loss": 0.4618,
|
| 1229 |
+
"step": 63000
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 2.09,
|
| 1233 |
+
"learning_rate": 9.090519938094768e-06,
|
| 1234 |
+
"loss": 0.4942,
|
| 1235 |
+
"step": 63500
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 2.11,
|
| 1239 |
+
"learning_rate": 8.925878362804176e-06,
|
| 1240 |
+
"loss": 0.4685,
|
| 1241 |
+
"step": 64000
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.11,
|
| 1245 |
+
"eval_exact_match": 5.4,
|
| 1246 |
+
"eval_f1": 23.1173,
|
| 1247 |
+
"eval_gen_len": 5.868,
|
| 1248 |
+
"eval_loss": 0.5359737277030945,
|
| 1249 |
+
"eval_rouge1": 28.1371,
|
| 1250 |
+
"eval_rouge2": 12.7219,
|
| 1251 |
+
"eval_rougeL": 28.0627,
|
| 1252 |
+
"eval_rougeLsum": 28.1156,
|
| 1253 |
+
"eval_runtime": 77.0613,
|
| 1254 |
+
"eval_samples_per_second": 12.977,
|
| 1255 |
+
"eval_steps_per_second": 4.334,
|
| 1256 |
+
"step": 64000
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 2.12,
|
| 1260 |
+
"learning_rate": 8.761236787513582e-06,
|
| 1261 |
+
"loss": 0.4914,
|
| 1262 |
+
"step": 64500
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 2.14,
|
| 1266 |
+
"learning_rate": 8.59659521222299e-06,
|
| 1267 |
+
"loss": 0.519,
|
| 1268 |
+
"step": 65000
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 2.16,
|
| 1272 |
+
"learning_rate": 8.4319536369324e-06,
|
| 1273 |
+
"loss": 0.4865,
|
| 1274 |
+
"step": 65500
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 2.17,
|
| 1278 |
+
"learning_rate": 8.267312061641807e-06,
|
| 1279 |
+
"loss": 0.4889,
|
| 1280 |
+
"step": 66000
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 2.17,
|
| 1284 |
+
"eval_exact_match": 5.8,
|
| 1285 |
+
"eval_f1": 23.0678,
|
| 1286 |
+
"eval_gen_len": 5.791,
|
| 1287 |
+
"eval_loss": 0.550064206123352,
|
| 1288 |
+
"eval_rouge1": 28.0394,
|
| 1289 |
+
"eval_rouge2": 12.3157,
|
| 1290 |
+
"eval_rougeL": 27.9714,
|
| 1291 |
+
"eval_rougeLsum": 27.9793,
|
| 1292 |
+
"eval_runtime": 76.4779,
|
| 1293 |
+
"eval_samples_per_second": 13.076,
|
| 1294 |
+
"eval_steps_per_second": 4.367,
|
| 1295 |
+
"step": 66000
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 2.19,
|
| 1299 |
+
"learning_rate": 8.102670486351213e-06,
|
| 1300 |
+
"loss": 0.4684,
|
| 1301 |
+
"step": 66500
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 2.21,
|
| 1305 |
+
"learning_rate": 7.938028911060621e-06,
|
| 1306 |
+
"loss": 0.4699,
|
| 1307 |
+
"step": 67000
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 2.22,
|
| 1311 |
+
"learning_rate": 7.773387335770029e-06,
|
| 1312 |
+
"loss": 0.5216,
|
| 1313 |
+
"step": 67500
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 2.24,
|
| 1317 |
+
"learning_rate": 7.608745760479437e-06,
|
| 1318 |
+
"loss": 0.4384,
|
| 1319 |
+
"step": 68000
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 2.24,
|
| 1323 |
+
"eval_exact_match": 5.4,
|
| 1324 |
+
"eval_f1": 23.1485,
|
| 1325 |
+
"eval_gen_len": 5.799,
|
| 1326 |
+
"eval_loss": 0.5489073991775513,
|
| 1327 |
+
"eval_rouge1": 28.3405,
|
| 1328 |
+
"eval_rouge2": 12.6721,
|
| 1329 |
+
"eval_rougeL": 28.3062,
|
| 1330 |
+
"eval_rougeLsum": 28.3124,
|
| 1331 |
+
"eval_runtime": 76.6915,
|
| 1332 |
+
"eval_samples_per_second": 13.039,
|
| 1333 |
+
"eval_steps_per_second": 4.355,
|
| 1334 |
+
"step": 68000
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 2.26,
|
| 1338 |
+
"learning_rate": 7.444104185188844e-06,
|
| 1339 |
+
"loss": 0.48,
|
| 1340 |
+
"step": 68500
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 2.27,
|
| 1344 |
+
"learning_rate": 7.279462609898251e-06,
|
| 1345 |
+
"loss": 0.4582,
|
| 1346 |
+
"step": 69000
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 2.29,
|
| 1350 |
+
"learning_rate": 7.11482103460766e-06,
|
| 1351 |
+
"loss": 0.4575,
|
| 1352 |
+
"step": 69500
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 2.3,
|
| 1356 |
+
"learning_rate": 6.9501794593170666e-06,
|
| 1357 |
+
"loss": 0.4588,
|
| 1358 |
+
"step": 70000
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 2.3,
|
| 1362 |
+
"eval_exact_match": 5.6,
|
| 1363 |
+
"eval_f1": 23.2198,
|
| 1364 |
+
"eval_gen_len": 5.809,
|
| 1365 |
+
"eval_loss": 0.5535086393356323,
|
| 1366 |
+
"eval_rouge1": 28.2308,
|
| 1367 |
+
"eval_rouge2": 12.4573,
|
| 1368 |
+
"eval_rougeL": 28.1886,
|
| 1369 |
+
"eval_rougeLsum": 28.1639,
|
| 1370 |
+
"eval_runtime": 76.7297,
|
| 1371 |
+
"eval_samples_per_second": 13.033,
|
| 1372 |
+
"eval_steps_per_second": 4.353,
|
| 1373 |
+
"step": 70000
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 2.32,
|
| 1377 |
+
"learning_rate": 6.785537884026475e-06,
|
| 1378 |
+
"loss": 0.4737,
|
| 1379 |
+
"step": 70500
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 2.34,
|
| 1383 |
+
"learning_rate": 6.620896308735882e-06,
|
| 1384 |
+
"loss": 0.4799,
|
| 1385 |
+
"step": 71000
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 2.35,
|
| 1389 |
+
"learning_rate": 6.45625473344529e-06,
|
| 1390 |
+
"loss": 0.4545,
|
| 1391 |
+
"step": 71500
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 2.37,
|
| 1395 |
+
"learning_rate": 6.2916131581546976e-06,
|
| 1396 |
+
"loss": 0.5076,
|
| 1397 |
+
"step": 72000
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 2.37,
|
| 1401 |
+
"eval_exact_match": 6.7,
|
| 1402 |
+
"eval_f1": 22.6072,
|
| 1403 |
+
"eval_gen_len": 5.739,
|
| 1404 |
+
"eval_loss": 0.5515512824058533,
|
| 1405 |
+
"eval_rouge1": 28.0709,
|
| 1406 |
+
"eval_rouge2": 12.4522,
|
| 1407 |
+
"eval_rougeL": 27.9586,
|
| 1408 |
+
"eval_rougeLsum": 27.9825,
|
| 1409 |
+
"eval_runtime": 76.3124,
|
| 1410 |
+
"eval_samples_per_second": 13.104,
|
| 1411 |
+
"eval_steps_per_second": 4.377,
|
| 1412 |
+
"step": 72000
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 2.39,
|
| 1416 |
+
"learning_rate": 6.126971582864105e-06,
|
| 1417 |
+
"loss": 0.442,
|
| 1418 |
+
"step": 72500
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 2.4,
|
| 1422 |
+
"learning_rate": 5.962330007573513e-06,
|
| 1423 |
+
"loss": 0.4464,
|
| 1424 |
+
"step": 73000
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 2.42,
|
| 1428 |
+
"learning_rate": 5.79768843228292e-06,
|
| 1429 |
+
"loss": 0.4756,
|
| 1430 |
+
"step": 73500
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 2.44,
|
| 1434 |
+
"learning_rate": 5.633046856992328e-06,
|
| 1435 |
+
"loss": 0.4467,
|
| 1436 |
+
"step": 74000
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 2.44,
|
| 1440 |
+
"eval_exact_match": 5.2,
|
| 1441 |
+
"eval_f1": 22.9357,
|
| 1442 |
+
"eval_gen_len": 5.876,
|
| 1443 |
+
"eval_loss": 0.5473840832710266,
|
| 1444 |
+
"eval_rouge1": 28.2468,
|
| 1445 |
+
"eval_rouge2": 12.4971,
|
| 1446 |
+
"eval_rougeL": 28.1601,
|
| 1447 |
+
"eval_rougeLsum": 28.1898,
|
| 1448 |
+
"eval_runtime": 77.1004,
|
| 1449 |
+
"eval_samples_per_second": 12.97,
|
| 1450 |
+
"eval_steps_per_second": 4.332,
|
| 1451 |
+
"step": 74000
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 2.45,
|
| 1455 |
+
"learning_rate": 5.4684052817017355e-06,
|
| 1456 |
+
"loss": 0.5136,
|
| 1457 |
+
"step": 74500
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 2.47,
|
| 1461 |
+
"learning_rate": 5.303763706411143e-06,
|
| 1462 |
+
"loss": 0.4626,
|
| 1463 |
+
"step": 75000
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 2.49,
|
| 1467 |
+
"learning_rate": 5.13912213112055e-06,
|
| 1468 |
+
"loss": 0.5138,
|
| 1469 |
+
"step": 75500
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 2.5,
|
| 1473 |
+
"learning_rate": 4.974480555829959e-06,
|
| 1474 |
+
"loss": 0.4262,
|
| 1475 |
+
"step": 76000
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 2.5,
|
| 1479 |
+
"eval_exact_match": 5.5,
|
| 1480 |
+
"eval_f1": 23.0092,
|
| 1481 |
+
"eval_gen_len": 5.883,
|
| 1482 |
+
"eval_loss": 0.5560488700866699,
|
| 1483 |
+
"eval_rouge1": 28.3505,
|
| 1484 |
+
"eval_rouge2": 12.633,
|
| 1485 |
+
"eval_rougeL": 28.2747,
|
| 1486 |
+
"eval_rougeLsum": 28.2859,
|
| 1487 |
+
"eval_runtime": 77.2929,
|
| 1488 |
+
"eval_samples_per_second": 12.938,
|
| 1489 |
+
"eval_steps_per_second": 4.321,
|
| 1490 |
+
"step": 76000
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 2.52,
|
| 1494 |
+
"learning_rate": 4.809838980539366e-06,
|
| 1495 |
+
"loss": 0.4773,
|
| 1496 |
+
"step": 76500
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 2.54,
|
| 1500 |
+
"learning_rate": 4.645197405248774e-06,
|
| 1501 |
+
"loss": 0.4813,
|
| 1502 |
+
"step": 77000
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 2.55,
|
| 1506 |
+
"learning_rate": 4.480555829958181e-06,
|
| 1507 |
+
"loss": 0.4833,
|
| 1508 |
+
"step": 77500
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 2.57,
|
| 1512 |
+
"learning_rate": 4.315914254667588e-06,
|
| 1513 |
+
"loss": 0.4812,
|
| 1514 |
+
"step": 78000
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 2.57,
|
| 1518 |
+
"eval_exact_match": 5.5,
|
| 1519 |
+
"eval_f1": 23.0871,
|
| 1520 |
+
"eval_gen_len": 5.812,
|
| 1521 |
+
"eval_loss": 0.5406239032745361,
|
| 1522 |
+
"eval_rouge1": 28.1686,
|
| 1523 |
+
"eval_rouge2": 12.4512,
|
| 1524 |
+
"eval_rougeL": 28.1366,
|
| 1525 |
+
"eval_rougeLsum": 28.1438,
|
| 1526 |
+
"eval_runtime": 76.6212,
|
| 1527 |
+
"eval_samples_per_second": 13.051,
|
| 1528 |
+
"eval_steps_per_second": 4.359,
|
| 1529 |
+
"step": 78000
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 2.58,
|
| 1533 |
+
"learning_rate": 4.151272679376997e-06,
|
| 1534 |
+
"loss": 0.4662,
|
| 1535 |
+
"step": 78500
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 2.6,
|
| 1539 |
+
"learning_rate": 3.9866311040864035e-06,
|
| 1540 |
+
"loss": 0.4939,
|
| 1541 |
+
"step": 79000
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 2.62,
|
| 1545 |
+
"learning_rate": 3.821989528795812e-06,
|
| 1546 |
+
"loss": 0.5221,
|
| 1547 |
+
"step": 79500
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 2.63,
|
| 1551 |
+
"learning_rate": 3.6573479535052195e-06,
|
| 1552 |
+
"loss": 0.4243,
|
| 1553 |
+
"step": 80000
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 2.63,
|
| 1557 |
+
"eval_exact_match": 5.2,
|
| 1558 |
+
"eval_f1": 22.7384,
|
| 1559 |
+
"eval_gen_len": 5.908,
|
| 1560 |
+
"eval_loss": 0.5526044964790344,
|
| 1561 |
+
"eval_rouge1": 28.0823,
|
| 1562 |
+
"eval_rouge2": 12.5204,
|
| 1563 |
+
"eval_rougeL": 28.0028,
|
| 1564 |
+
"eval_rougeLsum": 28.0533,
|
| 1565 |
+
"eval_runtime": 77.3998,
|
| 1566 |
+
"eval_samples_per_second": 12.92,
|
| 1567 |
+
"eval_steps_per_second": 4.315,
|
| 1568 |
+
"step": 80000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 2.65,
|
| 1572 |
+
"learning_rate": 3.492706378214627e-06,
|
| 1573 |
+
"loss": 0.4704,
|
| 1574 |
+
"step": 80500
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 2.67,
|
| 1578 |
+
"learning_rate": 3.328064802924034e-06,
|
| 1579 |
+
"loss": 0.4731,
|
| 1580 |
+
"step": 81000
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 2.68,
|
| 1584 |
+
"learning_rate": 3.163423227633442e-06,
|
| 1585 |
+
"loss": 0.4775,
|
| 1586 |
+
"step": 81500
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 2.7,
|
| 1590 |
+
"learning_rate": 2.9987816523428496e-06,
|
| 1591 |
+
"loss": 0.4828,
|
| 1592 |
+
"step": 82000
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 2.7,
|
| 1596 |
+
"eval_exact_match": 6.7,
|
| 1597 |
+
"eval_f1": 22.2536,
|
| 1598 |
+
"eval_gen_len": 5.75,
|
| 1599 |
+
"eval_loss": 0.5606855154037476,
|
| 1600 |
+
"eval_rouge1": 27.9073,
|
| 1601 |
+
"eval_rouge2": 12.4149,
|
| 1602 |
+
"eval_rougeL": 27.8389,
|
| 1603 |
+
"eval_rougeLsum": 27.8483,
|
| 1604 |
+
"eval_runtime": 76.3805,
|
| 1605 |
+
"eval_samples_per_second": 13.092,
|
| 1606 |
+
"eval_steps_per_second": 4.373,
|
| 1607 |
+
"step": 82000
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 2.72,
|
| 1611 |
+
"learning_rate": 2.8341400770522574e-06,
|
| 1612 |
+
"loss": 0.4479,
|
| 1613 |
+
"step": 82500
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 2.73,
|
| 1617 |
+
"learning_rate": 2.669498501761665e-06,
|
| 1618 |
+
"loss": 0.4923,
|
| 1619 |
+
"step": 83000
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 2.75,
|
| 1623 |
+
"learning_rate": 2.5048569264710725e-06,
|
| 1624 |
+
"loss": 0.5143,
|
| 1625 |
+
"step": 83500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 2.77,
|
| 1629 |
+
"learning_rate": 2.3402153511804802e-06,
|
| 1630 |
+
"loss": 0.4827,
|
| 1631 |
+
"step": 84000
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 2.77,
|
| 1635 |
+
"eval_exact_match": 4.8,
|
| 1636 |
+
"eval_f1": 23.1203,
|
| 1637 |
+
"eval_gen_len": 5.883,
|
| 1638 |
+
"eval_loss": 0.5355603694915771,
|
| 1639 |
+
"eval_rouge1": 28.5515,
|
| 1640 |
+
"eval_rouge2": 12.5974,
|
| 1641 |
+
"eval_rougeL": 28.4917,
|
| 1642 |
+
"eval_rougeLsum": 28.4884,
|
| 1643 |
+
"eval_runtime": 77.1315,
|
| 1644 |
+
"eval_samples_per_second": 12.965,
|
| 1645 |
+
"eval_steps_per_second": 4.33,
|
| 1646 |
+
"step": 84000
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 2.78,
|
| 1650 |
+
"learning_rate": 2.175573775889888e-06,
|
| 1651 |
+
"loss": 0.453,
|
| 1652 |
+
"step": 84500
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 2.8,
|
| 1656 |
+
"learning_rate": 2.0109322005992957e-06,
|
| 1657 |
+
"loss": 0.4689,
|
| 1658 |
+
"step": 85000
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 2.82,
|
| 1662 |
+
"learning_rate": 1.846290625308703e-06,
|
| 1663 |
+
"loss": 0.4628,
|
| 1664 |
+
"step": 85500
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 2.83,
|
| 1668 |
+
"learning_rate": 1.6816490500181106e-06,
|
| 1669 |
+
"loss": 0.4599,
|
| 1670 |
+
"step": 86000
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 2.83,
|
| 1674 |
+
"eval_exact_match": 6.0,
|
| 1675 |
+
"eval_f1": 22.775,
|
| 1676 |
+
"eval_gen_len": 5.75,
|
| 1677 |
+
"eval_loss": 0.5486857891082764,
|
| 1678 |
+
"eval_rouge1": 28.1728,
|
| 1679 |
+
"eval_rouge2": 12.3968,
|
| 1680 |
+
"eval_rougeL": 28.1105,
|
| 1681 |
+
"eval_rougeLsum": 28.1316,
|
| 1682 |
+
"eval_runtime": 76.2649,
|
| 1683 |
+
"eval_samples_per_second": 13.112,
|
| 1684 |
+
"eval_steps_per_second": 4.379,
|
| 1685 |
+
"step": 86000
|
| 1686 |
+
},
|
| 1687 |
+
{
|
| 1688 |
+
"epoch": 2.85,
|
| 1689 |
+
"learning_rate": 1.5170074747275183e-06,
|
| 1690 |
+
"loss": 0.4783,
|
| 1691 |
+
"step": 86500
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 2.86,
|
| 1695 |
+
"learning_rate": 1.3523658994369257e-06,
|
| 1696 |
+
"loss": 0.4793,
|
| 1697 |
+
"step": 87000
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 2.88,
|
| 1701 |
+
"learning_rate": 1.1877243241463334e-06,
|
| 1702 |
+
"loss": 0.4457,
|
| 1703 |
+
"step": 87500
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 2.9,
|
| 1707 |
+
"learning_rate": 1.0230827488557412e-06,
|
| 1708 |
+
"loss": 0.4505,
|
| 1709 |
+
"step": 88000
|
| 1710 |
+
},
|
| 1711 |
+
{
|
| 1712 |
+
"epoch": 2.9,
|
| 1713 |
+
"eval_exact_match": 5.3,
|
| 1714 |
+
"eval_f1": 23.063,
|
| 1715 |
+
"eval_gen_len": 5.832,
|
| 1716 |
+
"eval_loss": 0.5459786057472229,
|
| 1717 |
+
"eval_rouge1": 28.381,
|
| 1718 |
+
"eval_rouge2": 12.6567,
|
| 1719 |
+
"eval_rougeL": 28.3519,
|
| 1720 |
+
"eval_rougeLsum": 28.3424,
|
| 1721 |
+
"eval_runtime": 76.7382,
|
| 1722 |
+
"eval_samples_per_second": 13.031,
|
| 1723 |
+
"eval_steps_per_second": 4.352,
|
| 1724 |
+
"step": 88000
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 2.91,
|
| 1728 |
+
"learning_rate": 8.584411735651487e-07,
|
| 1729 |
+
"loss": 0.4594,
|
| 1730 |
+
"step": 88500
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 2.93,
|
| 1734 |
+
"learning_rate": 6.937995982745563e-07,
|
| 1735 |
+
"loss": 0.4719,
|
| 1736 |
+
"step": 89000
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 2.95,
|
| 1740 |
+
"learning_rate": 5.291580229839639e-07,
|
| 1741 |
+
"loss": 0.4538,
|
| 1742 |
+
"step": 89500
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 2.96,
|
| 1746 |
+
"learning_rate": 3.6451644769337155e-07,
|
| 1747 |
+
"loss": 0.4741,
|
| 1748 |
+
"step": 90000
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 2.96,
|
| 1752 |
+
"eval_exact_match": 5.3,
|
| 1753 |
+
"eval_f1": 23.0565,
|
| 1754 |
+
"eval_gen_len": 5.819,
|
| 1755 |
+
"eval_loss": 0.5445623993873596,
|
| 1756 |
+
"eval_rouge1": 28.2709,
|
| 1757 |
+
"eval_rouge2": 12.4988,
|
| 1758 |
+
"eval_rougeL": 28.1962,
|
| 1759 |
+
"eval_rougeLsum": 28.2203,
|
| 1760 |
+
"eval_runtime": 76.5919,
|
| 1761 |
+
"eval_samples_per_second": 13.056,
|
| 1762 |
+
"eval_steps_per_second": 4.361,
|
| 1763 |
+
"step": 90000
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 2.98,
|
| 1767 |
+
"learning_rate": 1.9987487240277914e-07,
|
| 1768 |
+
"loss": 0.441,
|
| 1769 |
+
"step": 90500
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 3.0,
|
| 1773 |
+
"learning_rate": 3.5233297112186775e-08,
|
| 1774 |
+
"loss": 0.4993,
|
| 1775 |
+
"step": 91000
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 3.0,
|
| 1779 |
+
"step": 91107,
|
| 1780 |
+
"total_flos": 1.4035529182858445e+17,
|
| 1781 |
+
"train_loss": 0.5681342689636171,
|
| 1782 |
+
"train_runtime": 25047.2432,
|
| 1783 |
+
"train_samples_per_second": 10.912,
|
| 1784 |
+
"train_steps_per_second": 3.637
|
| 1785 |
+
}
|
| 1786 |
+
],
|
| 1787 |
+
"max_steps": 91107,
|
| 1788 |
+
"num_train_epochs": 3,
|
| 1789 |
+
"total_flos": 1.4035529182858445e+17,
|
| 1790 |
+
"trial_name": null,
|
| 1791 |
+
"trial_params": null
|
| 1792 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21252bdb3ce263a8e09b679e42bf6a65ea56880856cccb16a62a6f1b23bc45a7
|
| 3 |
+
size 3183
|