jiang commited on
Commit ·
da480a1
1
Parent(s): 5cabd8e
First version
Browse files- config.json +56 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -0
- spiece.model +3 -0
- tokenizer_config.json +1 -0
- trainer_state.json +724 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "t5-small",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"T5ForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"d_ff": 2048,
|
| 7 |
+
"d_kv": 64,
|
| 8 |
+
"d_model": 512,
|
| 9 |
+
"decoder_start_token_id": 0,
|
| 10 |
+
"dropout_rate": 0.1,
|
| 11 |
+
"eos_token_id": 1,
|
| 12 |
+
"feed_forward_proj": "relu",
|
| 13 |
+
"initializer_factor": 1.0,
|
| 14 |
+
"is_encoder_decoder": true,
|
| 15 |
+
"layer_norm_epsilon": 1e-06,
|
| 16 |
+
"model_type": "t5",
|
| 17 |
+
"n_positions": 512,
|
| 18 |
+
"num_decoder_layers": 6,
|
| 19 |
+
"num_heads": 8,
|
| 20 |
+
"num_layers": 6,
|
| 21 |
+
"output_past": true,
|
| 22 |
+
"pad_token_id": 0,
|
| 23 |
+
"relative_attention_num_buckets": 32,
|
| 24 |
+
"task_specific_params": {
|
| 25 |
+
"summarization": {
|
| 26 |
+
"early_stopping": true,
|
| 27 |
+
"length_penalty": 2.0,
|
| 28 |
+
"max_length": 200,
|
| 29 |
+
"min_length": 30,
|
| 30 |
+
"no_repeat_ngram_size": 3,
|
| 31 |
+
"num_beams": 4,
|
| 32 |
+
"prefix": "summarize: "
|
| 33 |
+
},
|
| 34 |
+
"translation_en_to_de": {
|
| 35 |
+
"early_stopping": true,
|
| 36 |
+
"max_length": 300,
|
| 37 |
+
"num_beams": 4,
|
| 38 |
+
"prefix": "translate English to German: "
|
| 39 |
+
},
|
| 40 |
+
"translation_en_to_fr": {
|
| 41 |
+
"early_stopping": true,
|
| 42 |
+
"max_length": 300,
|
| 43 |
+
"num_beams": 4,
|
| 44 |
+
"prefix": "translate English to French: "
|
| 45 |
+
},
|
| 46 |
+
"translation_en_to_ro": {
|
| 47 |
+
"early_stopping": true,
|
| 48 |
+
"max_length": 300,
|
| 49 |
+
"num_beams": 4,
|
| 50 |
+
"prefix": "translate English to Romanian: "
|
| 51 |
+
}
|
| 52 |
+
},
|
| 53 |
+
"transformers_version": "4.5.1",
|
| 54 |
+
"use_cache": true,
|
| 55 |
+
"vocab_size": 32128
|
| 56 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f11ef9ef4fceccd1246471ee79078b665aeb23fb7ee714049126814a9d77b870
|
| 3 |
+
size 484136037
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3befab56ede24b4b3765bd4e838ae7616cd9e5e55ee47afd91b81e724acaa4a7
|
| 3 |
+
size 242089354
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:58a4d2a0b45b521bda693766a5ff757f1080096aa8affb3b51339a995487789c
|
| 3 |
+
size 623
|
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>"]}
|
spiece.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
|
| 3 |
+
size 791656
|
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>"], "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "t5-small"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,724 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 50000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.1,
|
| 12 |
+
"learning_rate": 1.98e-05,
|
| 13 |
+
"loss": 1.4142,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.2,
|
| 18 |
+
"learning_rate": 1.9600000000000002e-05,
|
| 19 |
+
"loss": 1.2006,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.3,
|
| 24 |
+
"learning_rate": 1.94e-05,
|
| 25 |
+
"loss": 1.1483,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.4,
|
| 30 |
+
"learning_rate": 1.9200000000000003e-05,
|
| 31 |
+
"loss": 1.1203,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.5,
|
| 36 |
+
"learning_rate": 1.9e-05,
|
| 37 |
+
"loss": 1.1042,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.6,
|
| 42 |
+
"learning_rate": 1.88e-05,
|
| 43 |
+
"loss": 1.0782,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.7,
|
| 48 |
+
"learning_rate": 1.86e-05,
|
| 49 |
+
"loss": 1.0579,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.8,
|
| 54 |
+
"learning_rate": 1.8400000000000003e-05,
|
| 55 |
+
"loss": 1.0638,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.9,
|
| 60 |
+
"learning_rate": 1.8200000000000002e-05,
|
| 61 |
+
"loss": 1.0532,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.0,
|
| 66 |
+
"learning_rate": 1.8e-05,
|
| 67 |
+
"loss": 1.0355,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.0,
|
| 72 |
+
"eval_gen_len": 9.6717,
|
| 73 |
+
"eval_loss": 0.9534130692481995,
|
| 74 |
+
"eval_rouge1": 11.1402,
|
| 75 |
+
"eval_rouge2": 2.3949,
|
| 76 |
+
"eval_rougeL": 11.0659,
|
| 77 |
+
"eval_rougeLsum": 11.1028,
|
| 78 |
+
"eval_runtime": 604.1787,
|
| 79 |
+
"eval_samples_per_second": 16.551,
|
| 80 |
+
"step": 5000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.1,
|
| 84 |
+
"learning_rate": 1.7800000000000002e-05,
|
| 85 |
+
"loss": 1.0246,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.2,
|
| 90 |
+
"learning_rate": 1.76e-05,
|
| 91 |
+
"loss": 1.0355,
|
| 92 |
+
"step": 6000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.3,
|
| 96 |
+
"learning_rate": 1.7400000000000003e-05,
|
| 97 |
+
"loss": 1.0076,
|
| 98 |
+
"step": 6500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.4,
|
| 102 |
+
"learning_rate": 1.72e-05,
|
| 103 |
+
"loss": 0.9938,
|
| 104 |
+
"step": 7000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.5,
|
| 108 |
+
"learning_rate": 1.7e-05,
|
| 109 |
+
"loss": 0.9928,
|
| 110 |
+
"step": 7500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 1.6,
|
| 114 |
+
"learning_rate": 1.6800000000000002e-05,
|
| 115 |
+
"loss": 1.0041,
|
| 116 |
+
"step": 8000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.7,
|
| 120 |
+
"learning_rate": 1.66e-05,
|
| 121 |
+
"loss": 0.9988,
|
| 122 |
+
"step": 8500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.8,
|
| 126 |
+
"learning_rate": 1.64e-05,
|
| 127 |
+
"loss": 0.9949,
|
| 128 |
+
"step": 9000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.9,
|
| 132 |
+
"learning_rate": 1.62e-05,
|
| 133 |
+
"loss": 0.9879,
|
| 134 |
+
"step": 9500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 2.0,
|
| 138 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 139 |
+
"loss": 1.0008,
|
| 140 |
+
"step": 10000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 2.0,
|
| 144 |
+
"eval_gen_len": 10.509,
|
| 145 |
+
"eval_loss": 0.9204540252685547,
|
| 146 |
+
"eval_rouge1": 11.224,
|
| 147 |
+
"eval_rouge2": 2.672,
|
| 148 |
+
"eval_rougeL": 11.1021,
|
| 149 |
+
"eval_rougeLsum": 11.1328,
|
| 150 |
+
"eval_runtime": 596.5212,
|
| 151 |
+
"eval_samples_per_second": 16.764,
|
| 152 |
+
"step": 10000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 2.1,
|
| 156 |
+
"learning_rate": 1.58e-05,
|
| 157 |
+
"loss": 0.9723,
|
| 158 |
+
"step": 10500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.2,
|
| 162 |
+
"learning_rate": 1.5600000000000003e-05,
|
| 163 |
+
"loss": 0.9784,
|
| 164 |
+
"step": 11000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 2.3,
|
| 168 |
+
"learning_rate": 1.54e-05,
|
| 169 |
+
"loss": 0.9629,
|
| 170 |
+
"step": 11500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 2.4,
|
| 174 |
+
"learning_rate": 1.5200000000000002e-05,
|
| 175 |
+
"loss": 0.9719,
|
| 176 |
+
"step": 12000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 2.5,
|
| 180 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 181 |
+
"loss": 0.9644,
|
| 182 |
+
"step": 12500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 2.6,
|
| 186 |
+
"learning_rate": 1.48e-05,
|
| 187 |
+
"loss": 0.9847,
|
| 188 |
+
"step": 13000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.7,
|
| 192 |
+
"learning_rate": 1.46e-05,
|
| 193 |
+
"loss": 0.9717,
|
| 194 |
+
"step": 13500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.8,
|
| 198 |
+
"learning_rate": 1.4400000000000001e-05,
|
| 199 |
+
"loss": 0.9644,
|
| 200 |
+
"step": 14000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.9,
|
| 204 |
+
"learning_rate": 1.4200000000000001e-05,
|
| 205 |
+
"loss": 0.9618,
|
| 206 |
+
"step": 14500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 3.0,
|
| 210 |
+
"learning_rate": 1.4e-05,
|
| 211 |
+
"loss": 0.9602,
|
| 212 |
+
"step": 15000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 3.0,
|
| 216 |
+
"eval_gen_len": 9.474,
|
| 217 |
+
"eval_loss": 0.9030117392539978,
|
| 218 |
+
"eval_rouge1": 12.4503,
|
| 219 |
+
"eval_rouge2": 2.8721,
|
| 220 |
+
"eval_rougeL": 12.354,
|
| 221 |
+
"eval_rougeLsum": 12.3602,
|
| 222 |
+
"eval_runtime": 585.0965,
|
| 223 |
+
"eval_samples_per_second": 17.091,
|
| 224 |
+
"step": 15000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 3.1,
|
| 228 |
+
"learning_rate": 1.38e-05,
|
| 229 |
+
"loss": 0.9477,
|
| 230 |
+
"step": 15500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 3.2,
|
| 234 |
+
"learning_rate": 1.3600000000000002e-05,
|
| 235 |
+
"loss": 0.9582,
|
| 236 |
+
"step": 16000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 3.3,
|
| 240 |
+
"learning_rate": 1.3400000000000002e-05,
|
| 241 |
+
"loss": 0.9391,
|
| 242 |
+
"step": 16500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 3.4,
|
| 246 |
+
"learning_rate": 1.3200000000000002e-05,
|
| 247 |
+
"loss": 0.9626,
|
| 248 |
+
"step": 17000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 3.5,
|
| 252 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 253 |
+
"loss": 0.951,
|
| 254 |
+
"step": 17500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 3.6,
|
| 258 |
+
"learning_rate": 1.2800000000000001e-05,
|
| 259 |
+
"loss": 0.9472,
|
| 260 |
+
"step": 18000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 3.7,
|
| 264 |
+
"learning_rate": 1.2600000000000001e-05,
|
| 265 |
+
"loss": 0.9519,
|
| 266 |
+
"step": 18500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 3.8,
|
| 270 |
+
"learning_rate": 1.2400000000000002e-05,
|
| 271 |
+
"loss": 0.952,
|
| 272 |
+
"step": 19000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 3.9,
|
| 276 |
+
"learning_rate": 1.22e-05,
|
| 277 |
+
"loss": 0.945,
|
| 278 |
+
"step": 19500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 4.0,
|
| 282 |
+
"learning_rate": 1.2e-05,
|
| 283 |
+
"loss": 0.9381,
|
| 284 |
+
"step": 20000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 4.0,
|
| 288 |
+
"eval_gen_len": 9.4554,
|
| 289 |
+
"eval_loss": 0.8888181447982788,
|
| 290 |
+
"eval_rouge1": 12.4117,
|
| 291 |
+
"eval_rouge2": 2.9965,
|
| 292 |
+
"eval_rougeL": 12.3285,
|
| 293 |
+
"eval_rougeLsum": 12.3327,
|
| 294 |
+
"eval_runtime": 592.8034,
|
| 295 |
+
"eval_samples_per_second": 16.869,
|
| 296 |
+
"step": 20000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 4.1,
|
| 300 |
+
"learning_rate": 1.18e-05,
|
| 301 |
+
"loss": 0.9535,
|
| 302 |
+
"step": 20500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 4.2,
|
| 306 |
+
"learning_rate": 1.16e-05,
|
| 307 |
+
"loss": 0.9458,
|
| 308 |
+
"step": 21000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 4.3,
|
| 312 |
+
"learning_rate": 1.14e-05,
|
| 313 |
+
"loss": 0.9321,
|
| 314 |
+
"step": 21500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 4.4,
|
| 318 |
+
"learning_rate": 1.1200000000000001e-05,
|
| 319 |
+
"loss": 0.9446,
|
| 320 |
+
"step": 22000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 4.5,
|
| 324 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 325 |
+
"loss": 0.9231,
|
| 326 |
+
"step": 22500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 4.6,
|
| 330 |
+
"learning_rate": 1.0800000000000002e-05,
|
| 331 |
+
"loss": 0.9285,
|
| 332 |
+
"step": 23000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 4.7,
|
| 336 |
+
"learning_rate": 1.0600000000000002e-05,
|
| 337 |
+
"loss": 0.9367,
|
| 338 |
+
"step": 23500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 4.8,
|
| 342 |
+
"learning_rate": 1.04e-05,
|
| 343 |
+
"loss": 0.9361,
|
| 344 |
+
"step": 24000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 4.9,
|
| 348 |
+
"learning_rate": 1.02e-05,
|
| 349 |
+
"loss": 0.9194,
|
| 350 |
+
"step": 24500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 5.0,
|
| 354 |
+
"learning_rate": 1e-05,
|
| 355 |
+
"loss": 0.9288,
|
| 356 |
+
"step": 25000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 5.0,
|
| 360 |
+
"eval_gen_len": 9.6762,
|
| 361 |
+
"eval_loss": 0.8804787993431091,
|
| 362 |
+
"eval_rouge1": 12.6469,
|
| 363 |
+
"eval_rouge2": 2.9175,
|
| 364 |
+
"eval_rougeL": 12.5557,
|
| 365 |
+
"eval_rougeLsum": 12.5805,
|
| 366 |
+
"eval_runtime": 595.0767,
|
| 367 |
+
"eval_samples_per_second": 16.805,
|
| 368 |
+
"step": 25000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 5.1,
|
| 372 |
+
"learning_rate": 9.800000000000001e-06,
|
| 373 |
+
"loss": 0.9299,
|
| 374 |
+
"step": 25500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 5.2,
|
| 378 |
+
"learning_rate": 9.600000000000001e-06,
|
| 379 |
+
"loss": 0.9313,
|
| 380 |
+
"step": 26000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 5.3,
|
| 384 |
+
"learning_rate": 9.4e-06,
|
| 385 |
+
"loss": 0.9239,
|
| 386 |
+
"step": 26500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 5.4,
|
| 390 |
+
"learning_rate": 9.200000000000002e-06,
|
| 391 |
+
"loss": 0.9238,
|
| 392 |
+
"step": 27000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 5.5,
|
| 396 |
+
"learning_rate": 9e-06,
|
| 397 |
+
"loss": 0.9291,
|
| 398 |
+
"step": 27500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 5.6,
|
| 402 |
+
"learning_rate": 8.8e-06,
|
| 403 |
+
"loss": 0.9207,
|
| 404 |
+
"step": 28000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 5.7,
|
| 408 |
+
"learning_rate": 8.6e-06,
|
| 409 |
+
"loss": 0.9256,
|
| 410 |
+
"step": 28500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 5.8,
|
| 414 |
+
"learning_rate": 8.400000000000001e-06,
|
| 415 |
+
"loss": 0.9199,
|
| 416 |
+
"step": 29000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 5.9,
|
| 420 |
+
"learning_rate": 8.2e-06,
|
| 421 |
+
"loss": 0.9113,
|
| 422 |
+
"step": 29500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 6.0,
|
| 426 |
+
"learning_rate": 8.000000000000001e-06,
|
| 427 |
+
"loss": 0.9243,
|
| 428 |
+
"step": 30000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 6.0,
|
| 432 |
+
"eval_gen_len": 10.0823,
|
| 433 |
+
"eval_loss": 0.8752232193946838,
|
| 434 |
+
"eval_rouge1": 14.0898,
|
| 435 |
+
"eval_rouge2": 3.4389,
|
| 436 |
+
"eval_rougeL": 13.9627,
|
| 437 |
+
"eval_rougeLsum": 13.9793,
|
| 438 |
+
"eval_runtime": 599.6266,
|
| 439 |
+
"eval_samples_per_second": 16.677,
|
| 440 |
+
"step": 30000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 6.1,
|
| 444 |
+
"learning_rate": 7.800000000000002e-06,
|
| 445 |
+
"loss": 0.9268,
|
| 446 |
+
"step": 30500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 6.2,
|
| 450 |
+
"learning_rate": 7.600000000000001e-06,
|
| 451 |
+
"loss": 0.9144,
|
| 452 |
+
"step": 31000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 6.3,
|
| 456 |
+
"learning_rate": 7.4e-06,
|
| 457 |
+
"loss": 0.9152,
|
| 458 |
+
"step": 31500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 6.4,
|
| 462 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 463 |
+
"loss": 0.9016,
|
| 464 |
+
"step": 32000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 6.5,
|
| 468 |
+
"learning_rate": 7e-06,
|
| 469 |
+
"loss": 0.9119,
|
| 470 |
+
"step": 32500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 6.6,
|
| 474 |
+
"learning_rate": 6.800000000000001e-06,
|
| 475 |
+
"loss": 0.916,
|
| 476 |
+
"step": 33000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 6.7,
|
| 480 |
+
"learning_rate": 6.600000000000001e-06,
|
| 481 |
+
"loss": 0.9253,
|
| 482 |
+
"step": 33500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 6.8,
|
| 486 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 487 |
+
"loss": 0.9199,
|
| 488 |
+
"step": 34000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 6.9,
|
| 492 |
+
"learning_rate": 6.200000000000001e-06,
|
| 493 |
+
"loss": 0.9168,
|
| 494 |
+
"step": 34500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 7.0,
|
| 498 |
+
"learning_rate": 6e-06,
|
| 499 |
+
"loss": 0.9087,
|
| 500 |
+
"step": 35000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 7.0,
|
| 504 |
+
"eval_gen_len": 9.7588,
|
| 505 |
+
"eval_loss": 0.8731149435043335,
|
| 506 |
+
"eval_rouge1": 13.9046,
|
| 507 |
+
"eval_rouge2": 3.2515,
|
| 508 |
+
"eval_rougeL": 13.7955,
|
| 509 |
+
"eval_rougeLsum": 13.8149,
|
| 510 |
+
"eval_runtime": 600.5144,
|
| 511 |
+
"eval_samples_per_second": 16.652,
|
| 512 |
+
"step": 35000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 7.1,
|
| 516 |
+
"learning_rate": 5.8e-06,
|
| 517 |
+
"loss": 0.905,
|
| 518 |
+
"step": 35500
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 7.2,
|
| 522 |
+
"learning_rate": 5.600000000000001e-06,
|
| 523 |
+
"loss": 0.9005,
|
| 524 |
+
"step": 36000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 7.3,
|
| 528 |
+
"learning_rate": 5.400000000000001e-06,
|
| 529 |
+
"loss": 0.9214,
|
| 530 |
+
"step": 36500
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 7.4,
|
| 534 |
+
"learning_rate": 5.2e-06,
|
| 535 |
+
"loss": 0.9152,
|
| 536 |
+
"step": 37000
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 7.5,
|
| 540 |
+
"learning_rate": 5e-06,
|
| 541 |
+
"loss": 0.9089,
|
| 542 |
+
"step": 37500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 7.6,
|
| 546 |
+
"learning_rate": 4.800000000000001e-06,
|
| 547 |
+
"loss": 0.9061,
|
| 548 |
+
"step": 38000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 7.7,
|
| 552 |
+
"learning_rate": 4.600000000000001e-06,
|
| 553 |
+
"loss": 0.918,
|
| 554 |
+
"step": 38500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 7.8,
|
| 558 |
+
"learning_rate": 4.4e-06,
|
| 559 |
+
"loss": 0.9055,
|
| 560 |
+
"step": 39000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 7.9,
|
| 564 |
+
"learning_rate": 4.2000000000000004e-06,
|
| 565 |
+
"loss": 0.9067,
|
| 566 |
+
"step": 39500
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 8.0,
|
| 570 |
+
"learning_rate": 4.000000000000001e-06,
|
| 571 |
+
"loss": 0.9146,
|
| 572 |
+
"step": 40000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 8.0,
|
| 576 |
+
"eval_gen_len": 9.8881,
|
| 577 |
+
"eval_loss": 0.871009886264801,
|
| 578 |
+
"eval_rouge1": 14.9652,
|
| 579 |
+
"eval_rouge2": 3.5015,
|
| 580 |
+
"eval_rougeL": 14.8443,
|
| 581 |
+
"eval_rougeLsum": 14.8507,
|
| 582 |
+
"eval_runtime": 606.6389,
|
| 583 |
+
"eval_samples_per_second": 16.484,
|
| 584 |
+
"step": 40000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 8.1,
|
| 588 |
+
"learning_rate": 3.8000000000000005e-06,
|
| 589 |
+
"loss": 0.9119,
|
| 590 |
+
"step": 40500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 8.2,
|
| 594 |
+
"learning_rate": 3.6000000000000003e-06,
|
| 595 |
+
"loss": 0.8942,
|
| 596 |
+
"step": 41000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 8.3,
|
| 600 |
+
"learning_rate": 3.4000000000000005e-06,
|
| 601 |
+
"loss": 0.9066,
|
| 602 |
+
"step": 41500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 8.4,
|
| 606 |
+
"learning_rate": 3.2000000000000003e-06,
|
| 607 |
+
"loss": 0.92,
|
| 608 |
+
"step": 42000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 8.5,
|
| 612 |
+
"learning_rate": 3e-06,
|
| 613 |
+
"loss": 0.9046,
|
| 614 |
+
"step": 42500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 8.6,
|
| 618 |
+
"learning_rate": 2.8000000000000003e-06,
|
| 619 |
+
"loss": 0.9027,
|
| 620 |
+
"step": 43000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 8.7,
|
| 624 |
+
"learning_rate": 2.6e-06,
|
| 625 |
+
"loss": 0.9019,
|
| 626 |
+
"step": 43500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 8.8,
|
| 630 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 631 |
+
"loss": 0.9056,
|
| 632 |
+
"step": 44000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 8.9,
|
| 636 |
+
"learning_rate": 2.2e-06,
|
| 637 |
+
"loss": 0.899,
|
| 638 |
+
"step": 44500
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 9.0,
|
| 642 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 643 |
+
"loss": 0.9034,
|
| 644 |
+
"step": 45000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 9.0,
|
| 648 |
+
"eval_gen_len": 9.8936,
|
| 649 |
+
"eval_loss": 0.8681530952453613,
|
| 650 |
+
"eval_rouge1": 14.9995,
|
| 651 |
+
"eval_rouge2": 3.5176,
|
| 652 |
+
"eval_rougeL": 14.8747,
|
| 653 |
+
"eval_rougeLsum": 14.8748,
|
| 654 |
+
"eval_runtime": 601.5997,
|
| 655 |
+
"eval_samples_per_second": 16.622,
|
| 656 |
+
"step": 45000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 9.1,
|
| 660 |
+
"learning_rate": 1.8000000000000001e-06,
|
| 661 |
+
"loss": 0.9079,
|
| 662 |
+
"step": 45500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 9.2,
|
| 666 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 667 |
+
"loss": 0.9027,
|
| 668 |
+
"step": 46000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 9.3,
|
| 672 |
+
"learning_rate": 1.4000000000000001e-06,
|
| 673 |
+
"loss": 0.8944,
|
| 674 |
+
"step": 46500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 9.4,
|
| 678 |
+
"learning_rate": 1.2000000000000002e-06,
|
| 679 |
+
"loss": 0.8955,
|
| 680 |
+
"step": 47000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 9.5,
|
| 684 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 685 |
+
"loss": 0.8949,
|
| 686 |
+
"step": 47500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 9.6,
|
| 690 |
+
"learning_rate": 8.000000000000001e-07,
|
| 691 |
+
"loss": 0.9037,
|
| 692 |
+
"step": 48000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 9.7,
|
| 696 |
+
"learning_rate": 6.000000000000001e-07,
|
| 697 |
+
"loss": 0.9074,
|
| 698 |
+
"step": 48500
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 9.8,
|
| 702 |
+
"learning_rate": 4.0000000000000003e-07,
|
| 703 |
+
"loss": 0.9145,
|
| 704 |
+
"step": 49000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 9.9,
|
| 708 |
+
"learning_rate": 2.0000000000000002e-07,
|
| 709 |
+
"loss": 0.9035,
|
| 710 |
+
"step": 49500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 10.0,
|
| 714 |
+
"learning_rate": 0.0,
|
| 715 |
+
"loss": 0.9147,
|
| 716 |
+
"step": 50000
|
| 717 |
+
}
|
| 718 |
+
],
|
| 719 |
+
"max_steps": 50000,
|
| 720 |
+
"num_train_epochs": 10,
|
| 721 |
+
"total_flos": 6.850590755050291e+16,
|
| 722 |
+
"trial_name": null,
|
| 723 |
+
"trial_params": null
|
| 724 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:76df6549c5b7becda291f98d08f5926ab0729b342bf06751840de0dbf1921784
|
| 3 |
+
size 2415
|