Training in progress, epoch 2, checkpoint
Browse files- checkpoint-102024/config.json +61 -0
- checkpoint-102024/generation_config.json +6 -0
- checkpoint-102024/optimizer.pt +3 -0
- checkpoint-102024/pytorch_model.bin +3 -0
- checkpoint-102024/rng_state.pth +3 -0
- checkpoint-102024/scheduler.pt +3 -0
- checkpoint-102024/special_tokens_map.json +107 -0
- checkpoint-102024/spiece.model +3 -0
- checkpoint-102024/tokenizer.json +0 -0
- checkpoint-102024/tokenizer_config.json +111 -0
- checkpoint-102024/trainer_state.json +1269 -0
- checkpoint-102024/training_args.bin +3 -0
checkpoint-102024/config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "t5-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"T5ForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"classifier_dropout": 0.0,
|
| 7 |
+
"d_ff": 3072,
|
| 8 |
+
"d_kv": 64,
|
| 9 |
+
"d_model": 768,
|
| 10 |
+
"decoder_start_token_id": 0,
|
| 11 |
+
"dense_act_fn": "relu",
|
| 12 |
+
"dropout_rate": 0.1,
|
| 13 |
+
"eos_token_id": 1,
|
| 14 |
+
"feed_forward_proj": "relu",
|
| 15 |
+
"initializer_factor": 1.0,
|
| 16 |
+
"is_encoder_decoder": true,
|
| 17 |
+
"is_gated_act": false,
|
| 18 |
+
"layer_norm_epsilon": 1e-06,
|
| 19 |
+
"model_type": "t5",
|
| 20 |
+
"n_positions": 512,
|
| 21 |
+
"num_decoder_layers": 12,
|
| 22 |
+
"num_heads": 12,
|
| 23 |
+
"num_layers": 12,
|
| 24 |
+
"output_past": true,
|
| 25 |
+
"pad_token_id": 0,
|
| 26 |
+
"relative_attention_max_distance": 128,
|
| 27 |
+
"relative_attention_num_buckets": 32,
|
| 28 |
+
"task_specific_params": {
|
| 29 |
+
"summarization": {
|
| 30 |
+
"early_stopping": true,
|
| 31 |
+
"length_penalty": 2.0,
|
| 32 |
+
"max_length": 200,
|
| 33 |
+
"min_length": 30,
|
| 34 |
+
"no_repeat_ngram_size": 3,
|
| 35 |
+
"num_beams": 4,
|
| 36 |
+
"prefix": "summarize: "
|
| 37 |
+
},
|
| 38 |
+
"translation_en_to_de": {
|
| 39 |
+
"early_stopping": true,
|
| 40 |
+
"max_length": 300,
|
| 41 |
+
"num_beams": 4,
|
| 42 |
+
"prefix": "translate English to German: "
|
| 43 |
+
},
|
| 44 |
+
"translation_en_to_fr": {
|
| 45 |
+
"early_stopping": true,
|
| 46 |
+
"max_length": 300,
|
| 47 |
+
"num_beams": 4,
|
| 48 |
+
"prefix": "translate English to French: "
|
| 49 |
+
},
|
| 50 |
+
"translation_en_to_ro": {
|
| 51 |
+
"early_stopping": true,
|
| 52 |
+
"max_length": 300,
|
| 53 |
+
"num_beams": 4,
|
| 54 |
+
"prefix": "translate English to Romanian: "
|
| 55 |
+
}
|
| 56 |
+
},
|
| 57 |
+
"torch_dtype": "float32",
|
| 58 |
+
"transformers_version": "4.33.3",
|
| 59 |
+
"use_cache": true,
|
| 60 |
+
"vocab_size": 32128
|
| 61 |
+
}
|
checkpoint-102024/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"decoder_start_token_id": 0,
|
| 3 |
+
"eos_token_id": 1,
|
| 4 |
+
"pad_token_id": 0,
|
| 5 |
+
"transformers_version": "4.33.3"
|
| 6 |
+
}
|
checkpoint-102024/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7b479ef62202107afddef3cdcb8beafbc29f5aaae2bee764bbbd05a6ed4aa896
|
| 3 |
+
size 1783444357
|
checkpoint-102024/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be39f676fc84c4452ede27db35c5b4cb4c8702b61a5f1243eda23ae5cbca62e3
|
| 3 |
+
size 891702929
|
checkpoint-102024/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:36e22d42574ab9d2c0c8efbbfe540ca23d331f25e8454f7b5ad4b1dc9de72e4a
|
| 3 |
+
size 14575
|
checkpoint-102024/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4d9b2eaf9748b1857f65c2842873ddd6979021eb704b012b69a3c724c30f30b6
|
| 3 |
+
size 627
|
checkpoint-102024/special_tokens_map.json
ADDED
|
@@ -0,0 +1,107 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<extra_id_0>",
|
| 4 |
+
"<extra_id_1>",
|
| 5 |
+
"<extra_id_2>",
|
| 6 |
+
"<extra_id_3>",
|
| 7 |
+
"<extra_id_4>",
|
| 8 |
+
"<extra_id_5>",
|
| 9 |
+
"<extra_id_6>",
|
| 10 |
+
"<extra_id_7>",
|
| 11 |
+
"<extra_id_8>",
|
| 12 |
+
"<extra_id_9>",
|
| 13 |
+
"<extra_id_10>",
|
| 14 |
+
"<extra_id_11>",
|
| 15 |
+
"<extra_id_12>",
|
| 16 |
+
"<extra_id_13>",
|
| 17 |
+
"<extra_id_14>",
|
| 18 |
+
"<extra_id_15>",
|
| 19 |
+
"<extra_id_16>",
|
| 20 |
+
"<extra_id_17>",
|
| 21 |
+
"<extra_id_18>",
|
| 22 |
+
"<extra_id_19>",
|
| 23 |
+
"<extra_id_20>",
|
| 24 |
+
"<extra_id_21>",
|
| 25 |
+
"<extra_id_22>",
|
| 26 |
+
"<extra_id_23>",
|
| 27 |
+
"<extra_id_24>",
|
| 28 |
+
"<extra_id_25>",
|
| 29 |
+
"<extra_id_26>",
|
| 30 |
+
"<extra_id_27>",
|
| 31 |
+
"<extra_id_28>",
|
| 32 |
+
"<extra_id_29>",
|
| 33 |
+
"<extra_id_30>",
|
| 34 |
+
"<extra_id_31>",
|
| 35 |
+
"<extra_id_32>",
|
| 36 |
+
"<extra_id_33>",
|
| 37 |
+
"<extra_id_34>",
|
| 38 |
+
"<extra_id_35>",
|
| 39 |
+
"<extra_id_36>",
|
| 40 |
+
"<extra_id_37>",
|
| 41 |
+
"<extra_id_38>",
|
| 42 |
+
"<extra_id_39>",
|
| 43 |
+
"<extra_id_40>",
|
| 44 |
+
"<extra_id_41>",
|
| 45 |
+
"<extra_id_42>",
|
| 46 |
+
"<extra_id_43>",
|
| 47 |
+
"<extra_id_44>",
|
| 48 |
+
"<extra_id_45>",
|
| 49 |
+
"<extra_id_46>",
|
| 50 |
+
"<extra_id_47>",
|
| 51 |
+
"<extra_id_48>",
|
| 52 |
+
"<extra_id_49>",
|
| 53 |
+
"<extra_id_50>",
|
| 54 |
+
"<extra_id_51>",
|
| 55 |
+
"<extra_id_52>",
|
| 56 |
+
"<extra_id_53>",
|
| 57 |
+
"<extra_id_54>",
|
| 58 |
+
"<extra_id_55>",
|
| 59 |
+
"<extra_id_56>",
|
| 60 |
+
"<extra_id_57>",
|
| 61 |
+
"<extra_id_58>",
|
| 62 |
+
"<extra_id_59>",
|
| 63 |
+
"<extra_id_60>",
|
| 64 |
+
"<extra_id_61>",
|
| 65 |
+
"<extra_id_62>",
|
| 66 |
+
"<extra_id_63>",
|
| 67 |
+
"<extra_id_64>",
|
| 68 |
+
"<extra_id_65>",
|
| 69 |
+
"<extra_id_66>",
|
| 70 |
+
"<extra_id_67>",
|
| 71 |
+
"<extra_id_68>",
|
| 72 |
+
"<extra_id_69>",
|
| 73 |
+
"<extra_id_70>",
|
| 74 |
+
"<extra_id_71>",
|
| 75 |
+
"<extra_id_72>",
|
| 76 |
+
"<extra_id_73>",
|
| 77 |
+
"<extra_id_74>",
|
| 78 |
+
"<extra_id_75>",
|
| 79 |
+
"<extra_id_76>",
|
| 80 |
+
"<extra_id_77>",
|
| 81 |
+
"<extra_id_78>",
|
| 82 |
+
"<extra_id_79>",
|
| 83 |
+
"<extra_id_80>",
|
| 84 |
+
"<extra_id_81>",
|
| 85 |
+
"<extra_id_82>",
|
| 86 |
+
"<extra_id_83>",
|
| 87 |
+
"<extra_id_84>",
|
| 88 |
+
"<extra_id_85>",
|
| 89 |
+
"<extra_id_86>",
|
| 90 |
+
"<extra_id_87>",
|
| 91 |
+
"<extra_id_88>",
|
| 92 |
+
"<extra_id_89>",
|
| 93 |
+
"<extra_id_90>",
|
| 94 |
+
"<extra_id_91>",
|
| 95 |
+
"<extra_id_92>",
|
| 96 |
+
"<extra_id_93>",
|
| 97 |
+
"<extra_id_94>",
|
| 98 |
+
"<extra_id_95>",
|
| 99 |
+
"<extra_id_96>",
|
| 100 |
+
"<extra_id_97>",
|
| 101 |
+
"<extra_id_98>",
|
| 102 |
+
"<extra_id_99>"
|
| 103 |
+
],
|
| 104 |
+
"eos_token": "</s>",
|
| 105 |
+
"pad_token": "<pad>",
|
| 106 |
+
"unk_token": "<unk>"
|
| 107 |
+
}
|
checkpoint-102024/spiece.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
|
| 3 |
+
size 791656
|
checkpoint-102024/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-102024/tokenizer_config.json
ADDED
|
@@ -0,0 +1,111 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<extra_id_0>",
|
| 4 |
+
"<extra_id_1>",
|
| 5 |
+
"<extra_id_2>",
|
| 6 |
+
"<extra_id_3>",
|
| 7 |
+
"<extra_id_4>",
|
| 8 |
+
"<extra_id_5>",
|
| 9 |
+
"<extra_id_6>",
|
| 10 |
+
"<extra_id_7>",
|
| 11 |
+
"<extra_id_8>",
|
| 12 |
+
"<extra_id_9>",
|
| 13 |
+
"<extra_id_10>",
|
| 14 |
+
"<extra_id_11>",
|
| 15 |
+
"<extra_id_12>",
|
| 16 |
+
"<extra_id_13>",
|
| 17 |
+
"<extra_id_14>",
|
| 18 |
+
"<extra_id_15>",
|
| 19 |
+
"<extra_id_16>",
|
| 20 |
+
"<extra_id_17>",
|
| 21 |
+
"<extra_id_18>",
|
| 22 |
+
"<extra_id_19>",
|
| 23 |
+
"<extra_id_20>",
|
| 24 |
+
"<extra_id_21>",
|
| 25 |
+
"<extra_id_22>",
|
| 26 |
+
"<extra_id_23>",
|
| 27 |
+
"<extra_id_24>",
|
| 28 |
+
"<extra_id_25>",
|
| 29 |
+
"<extra_id_26>",
|
| 30 |
+
"<extra_id_27>",
|
| 31 |
+
"<extra_id_28>",
|
| 32 |
+
"<extra_id_29>",
|
| 33 |
+
"<extra_id_30>",
|
| 34 |
+
"<extra_id_31>",
|
| 35 |
+
"<extra_id_32>",
|
| 36 |
+
"<extra_id_33>",
|
| 37 |
+
"<extra_id_34>",
|
| 38 |
+
"<extra_id_35>",
|
| 39 |
+
"<extra_id_36>",
|
| 40 |
+
"<extra_id_37>",
|
| 41 |
+
"<extra_id_38>",
|
| 42 |
+
"<extra_id_39>",
|
| 43 |
+
"<extra_id_40>",
|
| 44 |
+
"<extra_id_41>",
|
| 45 |
+
"<extra_id_42>",
|
| 46 |
+
"<extra_id_43>",
|
| 47 |
+
"<extra_id_44>",
|
| 48 |
+
"<extra_id_45>",
|
| 49 |
+
"<extra_id_46>",
|
| 50 |
+
"<extra_id_47>",
|
| 51 |
+
"<extra_id_48>",
|
| 52 |
+
"<extra_id_49>",
|
| 53 |
+
"<extra_id_50>",
|
| 54 |
+
"<extra_id_51>",
|
| 55 |
+
"<extra_id_52>",
|
| 56 |
+
"<extra_id_53>",
|
| 57 |
+
"<extra_id_54>",
|
| 58 |
+
"<extra_id_55>",
|
| 59 |
+
"<extra_id_56>",
|
| 60 |
+
"<extra_id_57>",
|
| 61 |
+
"<extra_id_58>",
|
| 62 |
+
"<extra_id_59>",
|
| 63 |
+
"<extra_id_60>",
|
| 64 |
+
"<extra_id_61>",
|
| 65 |
+
"<extra_id_62>",
|
| 66 |
+
"<extra_id_63>",
|
| 67 |
+
"<extra_id_64>",
|
| 68 |
+
"<extra_id_65>",
|
| 69 |
+
"<extra_id_66>",
|
| 70 |
+
"<extra_id_67>",
|
| 71 |
+
"<extra_id_68>",
|
| 72 |
+
"<extra_id_69>",
|
| 73 |
+
"<extra_id_70>",
|
| 74 |
+
"<extra_id_71>",
|
| 75 |
+
"<extra_id_72>",
|
| 76 |
+
"<extra_id_73>",
|
| 77 |
+
"<extra_id_74>",
|
| 78 |
+
"<extra_id_75>",
|
| 79 |
+
"<extra_id_76>",
|
| 80 |
+
"<extra_id_77>",
|
| 81 |
+
"<extra_id_78>",
|
| 82 |
+
"<extra_id_79>",
|
| 83 |
+
"<extra_id_80>",
|
| 84 |
+
"<extra_id_81>",
|
| 85 |
+
"<extra_id_82>",
|
| 86 |
+
"<extra_id_83>",
|
| 87 |
+
"<extra_id_84>",
|
| 88 |
+
"<extra_id_85>",
|
| 89 |
+
"<extra_id_86>",
|
| 90 |
+
"<extra_id_87>",
|
| 91 |
+
"<extra_id_88>",
|
| 92 |
+
"<extra_id_89>",
|
| 93 |
+
"<extra_id_90>",
|
| 94 |
+
"<extra_id_91>",
|
| 95 |
+
"<extra_id_92>",
|
| 96 |
+
"<extra_id_93>",
|
| 97 |
+
"<extra_id_94>",
|
| 98 |
+
"<extra_id_95>",
|
| 99 |
+
"<extra_id_96>",
|
| 100 |
+
"<extra_id_97>",
|
| 101 |
+
"<extra_id_98>",
|
| 102 |
+
"<extra_id_99>"
|
| 103 |
+
],
|
| 104 |
+
"clean_up_tokenization_spaces": true,
|
| 105 |
+
"eos_token": "</s>",
|
| 106 |
+
"extra_ids": 100,
|
| 107 |
+
"model_max_length": 512,
|
| 108 |
+
"pad_token": "<pad>",
|
| 109 |
+
"tokenizer_class": "T5Tokenizer",
|
| 110 |
+
"unk_token": "<unk>"
|
| 111 |
+
}
|
checkpoint-102024/trainer_state.json
ADDED
|
@@ -0,0 +1,1269 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.1877,
|
| 3 |
+
"best_model_checkpoint": "xsum_finetuned_on_train2/checkpoint-102024",
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 102024,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01,
|
| 13 |
+
"learning_rate": 1.999805535952325e-05,
|
| 14 |
+
"loss": 2.4175,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.02,
|
| 19 |
+
"learning_rate": 1.9996106798400377e-05,
|
| 20 |
+
"loss": 2.2883,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.03,
|
| 25 |
+
"learning_rate": 1.9994162157923627e-05,
|
| 26 |
+
"loss": 2.2417,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.04,
|
| 31 |
+
"learning_rate": 1.9992209676154634e-05,
|
| 32 |
+
"loss": 2.2231,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.05,
|
| 37 |
+
"learning_rate": 1.999026503567788e-05,
|
| 38 |
+
"loss": 2.2194,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.06,
|
| 43 |
+
"learning_rate": 1.9988316474555006e-05,
|
| 44 |
+
"loss": 2.2268,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.07,
|
| 49 |
+
"learning_rate": 1.9986363992786014e-05,
|
| 50 |
+
"loss": 2.3025,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.08,
|
| 55 |
+
"learning_rate": 1.9984403669724773e-05,
|
| 56 |
+
"loss": 2.7304,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.09,
|
| 61 |
+
"learning_rate": 1.9982443346663532e-05,
|
| 62 |
+
"loss": 3.5109,
|
| 63 |
+
"step": 4500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.1,
|
| 67 |
+
"learning_rate": 1.9980490864894537e-05,
|
| 68 |
+
"loss": 4.2693,
|
| 69 |
+
"step": 5000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.11,
|
| 73 |
+
"learning_rate": 1.9978534462479418e-05,
|
| 74 |
+
"loss": 4.4567,
|
| 75 |
+
"step": 5500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.12,
|
| 79 |
+
"learning_rate": 1.9976574139418177e-05,
|
| 80 |
+
"loss": 4.3819,
|
| 81 |
+
"step": 6000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.13,
|
| 85 |
+
"learning_rate": 1.997462165764918e-05,
|
| 86 |
+
"loss": 4.2867,
|
| 87 |
+
"step": 6500
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.14,
|
| 91 |
+
"learning_rate": 1.9972665255234067e-05,
|
| 92 |
+
"loss": 4.2057,
|
| 93 |
+
"step": 7000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.15,
|
| 97 |
+
"learning_rate": 1.9970704932172826e-05,
|
| 98 |
+
"loss": 4.1772,
|
| 99 |
+
"step": 7500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.16,
|
| 103 |
+
"learning_rate": 1.9968748529757707e-05,
|
| 104 |
+
"loss": 4.1272,
|
| 105 |
+
"step": 8000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.17,
|
| 109 |
+
"learning_rate": 1.996679604798871e-05,
|
| 110 |
+
"loss": 4.0761,
|
| 111 |
+
"step": 8500
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.18,
|
| 115 |
+
"learning_rate": 1.996483572492747e-05,
|
| 116 |
+
"loss": 4.0732,
|
| 117 |
+
"step": 9000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.19,
|
| 121 |
+
"learning_rate": 1.9962879322512353e-05,
|
| 122 |
+
"loss": 4.0824,
|
| 123 |
+
"step": 9500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.2,
|
| 127 |
+
"learning_rate": 1.9960918999451112e-05,
|
| 128 |
+
"loss": 4.062,
|
| 129 |
+
"step": 10000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.21,
|
| 133 |
+
"learning_rate": 1.995895867638987e-05,
|
| 134 |
+
"loss": 4.0745,
|
| 135 |
+
"step": 10500
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.22,
|
| 139 |
+
"learning_rate": 1.995699835332863e-05,
|
| 140 |
+
"loss": 4.0372,
|
| 141 |
+
"step": 11000
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.23,
|
| 145 |
+
"learning_rate": 1.9955041950913512e-05,
|
| 146 |
+
"loss": 4.0514,
|
| 147 |
+
"step": 11500
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.24,
|
| 151 |
+
"learning_rate": 1.995308162785227e-05,
|
| 152 |
+
"loss": 4.076,
|
| 153 |
+
"step": 12000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.25,
|
| 157 |
+
"learning_rate": 1.9951140908021646e-05,
|
| 158 |
+
"loss": 4.0588,
|
| 159 |
+
"step": 12500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.25,
|
| 163 |
+
"learning_rate": 1.9949184505606524e-05,
|
| 164 |
+
"loss": 4.0508,
|
| 165 |
+
"step": 13000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.26,
|
| 169 |
+
"learning_rate": 1.9947232023837532e-05,
|
| 170 |
+
"loss": 4.0852,
|
| 171 |
+
"step": 13500
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.27,
|
| 175 |
+
"learning_rate": 1.994528346271466e-05,
|
| 176 |
+
"loss": 4.0821,
|
| 177 |
+
"step": 14000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.28,
|
| 181 |
+
"learning_rate": 1.9943323139653418e-05,
|
| 182 |
+
"loss": 4.0715,
|
| 183 |
+
"step": 14500
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.29,
|
| 187 |
+
"learning_rate": 1.9941362816592177e-05,
|
| 188 |
+
"loss": 4.0786,
|
| 189 |
+
"step": 15000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.3,
|
| 193 |
+
"learning_rate": 1.993940641417706e-05,
|
| 194 |
+
"loss": 4.072,
|
| 195 |
+
"step": 15500
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.31,
|
| 199 |
+
"learning_rate": 1.9937446091115818e-05,
|
| 200 |
+
"loss": 4.0919,
|
| 201 |
+
"step": 16000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.32,
|
| 205 |
+
"learning_rate": 1.99354896887007e-05,
|
| 206 |
+
"loss": 4.0854,
|
| 207 |
+
"step": 16500
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.33,
|
| 211 |
+
"learning_rate": 1.993352936563946e-05,
|
| 212 |
+
"loss": 4.0786,
|
| 213 |
+
"step": 17000
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.34,
|
| 217 |
+
"learning_rate": 1.993157296322434e-05,
|
| 218 |
+
"loss": 4.0864,
|
| 219 |
+
"step": 17500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.35,
|
| 223 |
+
"learning_rate": 1.9929616560809225e-05,
|
| 224 |
+
"loss": 4.0824,
|
| 225 |
+
"step": 18000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.36,
|
| 229 |
+
"learning_rate": 1.9927660158394104e-05,
|
| 230 |
+
"loss": 4.0776,
|
| 231 |
+
"step": 18500
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.37,
|
| 235 |
+
"learning_rate": 1.9925699835332863e-05,
|
| 236 |
+
"loss": 4.074,
|
| 237 |
+
"step": 19000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.38,
|
| 241 |
+
"learning_rate": 1.992374735356387e-05,
|
| 242 |
+
"loss": 4.1029,
|
| 243 |
+
"step": 19500
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.39,
|
| 247 |
+
"learning_rate": 1.992178703050263e-05,
|
| 248 |
+
"loss": 4.0631,
|
| 249 |
+
"step": 20000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.4,
|
| 253 |
+
"learning_rate": 1.991982670744139e-05,
|
| 254 |
+
"loss": 4.085,
|
| 255 |
+
"step": 20500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.41,
|
| 259 |
+
"learning_rate": 1.9917866384380148e-05,
|
| 260 |
+
"loss": 4.0807,
|
| 261 |
+
"step": 21000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.42,
|
| 265 |
+
"learning_rate": 1.991590998196503e-05,
|
| 266 |
+
"loss": 4.0622,
|
| 267 |
+
"step": 21500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.43,
|
| 271 |
+
"learning_rate": 1.9913957500196034e-05,
|
| 272 |
+
"loss": 4.0605,
|
| 273 |
+
"step": 22000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.44,
|
| 277 |
+
"learning_rate": 1.9911997177134793e-05,
|
| 278 |
+
"loss": 4.0597,
|
| 279 |
+
"step": 22500
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.45,
|
| 283 |
+
"learning_rate": 1.9910036854073552e-05,
|
| 284 |
+
"loss": 4.0646,
|
| 285 |
+
"step": 23000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.46,
|
| 289 |
+
"learning_rate": 1.9908084372304556e-05,
|
| 290 |
+
"loss": 4.0652,
|
| 291 |
+
"step": 23500
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 0.47,
|
| 295 |
+
"learning_rate": 1.9906127969889438e-05,
|
| 296 |
+
"loss": 4.0453,
|
| 297 |
+
"step": 24000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.48,
|
| 301 |
+
"learning_rate": 1.9904167646828197e-05,
|
| 302 |
+
"loss": 4.0989,
|
| 303 |
+
"step": 24500
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.49,
|
| 307 |
+
"learning_rate": 1.9902211244413082e-05,
|
| 308 |
+
"loss": 4.0759,
|
| 309 |
+
"step": 25000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.5,
|
| 313 |
+
"learning_rate": 1.990025092135184e-05,
|
| 314 |
+
"loss": 4.0907,
|
| 315 |
+
"step": 25500
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 0.51,
|
| 319 |
+
"learning_rate": 1.98982905982906e-05,
|
| 320 |
+
"loss": 4.0717,
|
| 321 |
+
"step": 26000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.52,
|
| 325 |
+
"learning_rate": 1.989633027522936e-05,
|
| 326 |
+
"loss": 4.0665,
|
| 327 |
+
"step": 26500
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.53,
|
| 331 |
+
"learning_rate": 1.989436995216812e-05,
|
| 332 |
+
"loss": 4.0185,
|
| 333 |
+
"step": 27000
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.54,
|
| 337 |
+
"learning_rate": 1.9892409629106878e-05,
|
| 338 |
+
"loss": 4.0697,
|
| 339 |
+
"step": 27500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.55,
|
| 343 |
+
"learning_rate": 1.9890449306045637e-05,
|
| 344 |
+
"loss": 4.0854,
|
| 345 |
+
"step": 28000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.56,
|
| 349 |
+
"learning_rate": 1.9888492903630522e-05,
|
| 350 |
+
"loss": 4.0598,
|
| 351 |
+
"step": 28500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 0.57,
|
| 355 |
+
"learning_rate": 1.9886532580569278e-05,
|
| 356 |
+
"loss": 4.0977,
|
| 357 |
+
"step": 29000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 0.58,
|
| 361 |
+
"learning_rate": 1.9884576178154163e-05,
|
| 362 |
+
"loss": 4.047,
|
| 363 |
+
"step": 29500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.59,
|
| 367 |
+
"learning_rate": 1.9882623696385167e-05,
|
| 368 |
+
"loss": 4.0893,
|
| 369 |
+
"step": 30000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.6,
|
| 373 |
+
"learning_rate": 1.9880663373323927e-05,
|
| 374 |
+
"loss": 4.07,
|
| 375 |
+
"step": 30500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.61,
|
| 379 |
+
"learning_rate": 1.9878703050262686e-05,
|
| 380 |
+
"loss": 4.0566,
|
| 381 |
+
"step": 31000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.62,
|
| 385 |
+
"learning_rate": 1.9876742727201445e-05,
|
| 386 |
+
"loss": 4.0784,
|
| 387 |
+
"step": 31500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.63,
|
| 391 |
+
"learning_rate": 1.9874782404140204e-05,
|
| 392 |
+
"loss": 4.0756,
|
| 393 |
+
"step": 32000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.64,
|
| 397 |
+
"learning_rate": 1.9872822081078963e-05,
|
| 398 |
+
"loss": 4.0628,
|
| 399 |
+
"step": 32500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.65,
|
| 403 |
+
"learning_rate": 1.9870865678663845e-05,
|
| 404 |
+
"loss": 4.0328,
|
| 405 |
+
"step": 33000
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.66,
|
| 409 |
+
"learning_rate": 1.9868909276248727e-05,
|
| 410 |
+
"loss": 4.0689,
|
| 411 |
+
"step": 33500
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 0.67,
|
| 415 |
+
"learning_rate": 1.986695679447973e-05,
|
| 416 |
+
"loss": 4.07,
|
| 417 |
+
"step": 34000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.68,
|
| 421 |
+
"learning_rate": 1.986499647141849e-05,
|
| 422 |
+
"loss": 4.0819,
|
| 423 |
+
"step": 34500
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.69,
|
| 427 |
+
"learning_rate": 1.9863040069003375e-05,
|
| 428 |
+
"loss": 4.0625,
|
| 429 |
+
"step": 35000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.7,
|
| 433 |
+
"learning_rate": 1.9861083666588254e-05,
|
| 434 |
+
"loss": 4.036,
|
| 435 |
+
"step": 35500
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 0.71,
|
| 439 |
+
"learning_rate": 1.9859123343527016e-05,
|
| 440 |
+
"loss": 4.068,
|
| 441 |
+
"step": 36000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.72,
|
| 445 |
+
"learning_rate": 1.9857163020465775e-05,
|
| 446 |
+
"loss": 4.0437,
|
| 447 |
+
"step": 36500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.73,
|
| 451 |
+
"learning_rate": 1.9855202697404534e-05,
|
| 452 |
+
"loss": 4.0635,
|
| 453 |
+
"step": 37000
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 0.74,
|
| 457 |
+
"learning_rate": 1.9853246294989416e-05,
|
| 458 |
+
"loss": 4.0766,
|
| 459 |
+
"step": 37500
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.74,
|
| 463 |
+
"learning_rate": 1.9851289892574298e-05,
|
| 464 |
+
"loss": 4.0651,
|
| 465 |
+
"step": 38000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.75,
|
| 469 |
+
"learning_rate": 1.9849329569513057e-05,
|
| 470 |
+
"loss": 4.044,
|
| 471 |
+
"step": 38500
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.76,
|
| 475 |
+
"learning_rate": 1.984737316709794e-05,
|
| 476 |
+
"loss": 4.0719,
|
| 477 |
+
"step": 39000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.77,
|
| 481 |
+
"learning_rate": 1.9845412844036698e-05,
|
| 482 |
+
"loss": 4.0501,
|
| 483 |
+
"step": 39500
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 0.78,
|
| 487 |
+
"learning_rate": 1.9843460362267702e-05,
|
| 488 |
+
"loss": 4.0683,
|
| 489 |
+
"step": 40000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.79,
|
| 493 |
+
"learning_rate": 1.984150003920646e-05,
|
| 494 |
+
"loss": 4.0753,
|
| 495 |
+
"step": 40500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 0.8,
|
| 499 |
+
"learning_rate": 1.983953971614522e-05,
|
| 500 |
+
"loss": 4.0873,
|
| 501 |
+
"step": 41000
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.81,
|
| 505 |
+
"learning_rate": 1.9837583313730106e-05,
|
| 506 |
+
"loss": 4.0842,
|
| 507 |
+
"step": 41500
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.82,
|
| 511 |
+
"learning_rate": 1.9835622990668865e-05,
|
| 512 |
+
"loss": 4.0603,
|
| 513 |
+
"step": 42000
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.83,
|
| 517 |
+
"learning_rate": 1.9833662667607624e-05,
|
| 518 |
+
"loss": 4.0556,
|
| 519 |
+
"step": 42500
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.84,
|
| 523 |
+
"learning_rate": 1.9831702344546383e-05,
|
| 524 |
+
"loss": 4.0557,
|
| 525 |
+
"step": 43000
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 0.85,
|
| 529 |
+
"learning_rate": 1.9829745942131265e-05,
|
| 530 |
+
"loss": 4.0726,
|
| 531 |
+
"step": 43500
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 0.86,
|
| 535 |
+
"learning_rate": 1.9827785619070024e-05,
|
| 536 |
+
"loss": 4.0716,
|
| 537 |
+
"step": 44000
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.87,
|
| 541 |
+
"learning_rate": 1.9825825296008787e-05,
|
| 542 |
+
"loss": 4.0514,
|
| 543 |
+
"step": 44500
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.88,
|
| 547 |
+
"learning_rate": 1.9823864972947546e-05,
|
| 548 |
+
"loss": 4.0772,
|
| 549 |
+
"step": 45000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.89,
|
| 553 |
+
"learning_rate": 1.9821904649886305e-05,
|
| 554 |
+
"loss": 4.0798,
|
| 555 |
+
"step": 45500
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.9,
|
| 559 |
+
"learning_rate": 1.9819948247471187e-05,
|
| 560 |
+
"loss": 4.0593,
|
| 561 |
+
"step": 46000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.91,
|
| 565 |
+
"learning_rate": 1.981799576570219e-05,
|
| 566 |
+
"loss": 4.0796,
|
| 567 |
+
"step": 46500
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 0.92,
|
| 571 |
+
"learning_rate": 1.981603544264095e-05,
|
| 572 |
+
"loss": 4.0398,
|
| 573 |
+
"step": 47000
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 0.93,
|
| 577 |
+
"learning_rate": 1.981407511957971e-05,
|
| 578 |
+
"loss": 4.0376,
|
| 579 |
+
"step": 47500
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.94,
|
| 583 |
+
"learning_rate": 1.981211479651847e-05,
|
| 584 |
+
"loss": 4.0384,
|
| 585 |
+
"step": 48000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.95,
|
| 589 |
+
"learning_rate": 1.9810154473457228e-05,
|
| 590 |
+
"loss": 4.0656,
|
| 591 |
+
"step": 48500
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.96,
|
| 595 |
+
"learning_rate": 1.980819807104211e-05,
|
| 596 |
+
"loss": 4.0767,
|
| 597 |
+
"step": 49000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.97,
|
| 601 |
+
"learning_rate": 1.980623774798087e-05,
|
| 602 |
+
"loss": 4.0442,
|
| 603 |
+
"step": 49500
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 0.98,
|
| 607 |
+
"learning_rate": 1.9804277424919628e-05,
|
| 608 |
+
"loss": 4.0622,
|
| 609 |
+
"step": 50000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.99,
|
| 613 |
+
"learning_rate": 1.9802317101858387e-05,
|
| 614 |
+
"loss": 4.0699,
|
| 615 |
+
"step": 50500
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 1.0,
|
| 619 |
+
"learning_rate": 1.9800360699443272e-05,
|
| 620 |
+
"loss": 4.051,
|
| 621 |
+
"step": 51000
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 1.0,
|
| 625 |
+
"eval_gen_len": 18.9831,
|
| 626 |
+
"eval_loss": 3.365176200866699,
|
| 627 |
+
"eval_rouge1": 0.1866,
|
| 628 |
+
"eval_rouge2": 0.0436,
|
| 629 |
+
"eval_rougeL": 0.1534,
|
| 630 |
+
"eval_rougeLsum": 0.1533,
|
| 631 |
+
"eval_runtime": 1349.6317,
|
| 632 |
+
"eval_samples_per_second": 8.396,
|
| 633 |
+
"eval_steps_per_second": 2.099,
|
| 634 |
+
"step": 51012
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 1.01,
|
| 638 |
+
"learning_rate": 1.979840429702815e-05,
|
| 639 |
+
"loss": 4.0511,
|
| 640 |
+
"step": 51500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.02,
|
| 644 |
+
"learning_rate": 1.9796443973966913e-05,
|
| 645 |
+
"loss": 4.0403,
|
| 646 |
+
"step": 52000
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.03,
|
| 650 |
+
"learning_rate": 1.9794483650905672e-05,
|
| 651 |
+
"loss": 4.0447,
|
| 652 |
+
"step": 52500
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 1.04,
|
| 656 |
+
"learning_rate": 1.979252332784443e-05,
|
| 657 |
+
"loss": 4.0273,
|
| 658 |
+
"step": 53000
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 1.05,
|
| 662 |
+
"learning_rate": 1.979056300478319e-05,
|
| 663 |
+
"loss": 4.0419,
|
| 664 |
+
"step": 53500
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 1.06,
|
| 668 |
+
"learning_rate": 1.978860268172195e-05,
|
| 669 |
+
"loss": 4.0677,
|
| 670 |
+
"step": 54000
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 1.07,
|
| 674 |
+
"learning_rate": 1.978664235866071e-05,
|
| 675 |
+
"loss": 4.0624,
|
| 676 |
+
"step": 54500
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.08,
|
| 680 |
+
"learning_rate": 1.978468595624559e-05,
|
| 681 |
+
"loss": 4.0649,
|
| 682 |
+
"step": 55000
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.09,
|
| 686 |
+
"learning_rate": 1.978272563318435e-05,
|
| 687 |
+
"loss": 4.0216,
|
| 688 |
+
"step": 55500
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.1,
|
| 692 |
+
"learning_rate": 1.978076923076923e-05,
|
| 693 |
+
"loss": 4.0691,
|
| 694 |
+
"step": 56000
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 1.11,
|
| 698 |
+
"learning_rate": 1.9778812828354116e-05,
|
| 699 |
+
"loss": 4.083,
|
| 700 |
+
"step": 56500
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 1.12,
|
| 704 |
+
"learning_rate": 1.9776856425938998e-05,
|
| 705 |
+
"loss": 4.0651,
|
| 706 |
+
"step": 57000
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 1.13,
|
| 710 |
+
"learning_rate": 1.9774900023523876e-05,
|
| 711 |
+
"loss": 4.0552,
|
| 712 |
+
"step": 57500
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 1.14,
|
| 716 |
+
"learning_rate": 1.977293970046264e-05,
|
| 717 |
+
"loss": 4.0682,
|
| 718 |
+
"step": 58000
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.15,
|
| 722 |
+
"learning_rate": 1.9770979377401398e-05,
|
| 723 |
+
"loss": 4.0457,
|
| 724 |
+
"step": 58500
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.16,
|
| 728 |
+
"learning_rate": 1.976902297498628e-05,
|
| 729 |
+
"loss": 4.0638,
|
| 730 |
+
"step": 59000
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.17,
|
| 734 |
+
"learning_rate": 1.976706265192504e-05,
|
| 735 |
+
"loss": 4.0493,
|
| 736 |
+
"step": 59500
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 1.18,
|
| 740 |
+
"learning_rate": 1.9765102328863798e-05,
|
| 741 |
+
"loss": 4.0795,
|
| 742 |
+
"step": 60000
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 1.19,
|
| 746 |
+
"learning_rate": 1.9763142005802557e-05,
|
| 747 |
+
"loss": 4.05,
|
| 748 |
+
"step": 60500
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 1.2,
|
| 752 |
+
"learning_rate": 1.9761181682741317e-05,
|
| 753 |
+
"loss": 4.0635,
|
| 754 |
+
"step": 61000
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 1.21,
|
| 758 |
+
"learning_rate": 1.9759221359680076e-05,
|
| 759 |
+
"loss": 4.0502,
|
| 760 |
+
"step": 61500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 1.22,
|
| 764 |
+
"learning_rate": 1.9757261036618835e-05,
|
| 765 |
+
"loss": 4.0559,
|
| 766 |
+
"step": 62000
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.23,
|
| 770 |
+
"learning_rate": 1.9755300713557597e-05,
|
| 771 |
+
"loss": 4.045,
|
| 772 |
+
"step": 62500
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.24,
|
| 776 |
+
"learning_rate": 1.9753352152434724e-05,
|
| 777 |
+
"loss": 4.0619,
|
| 778 |
+
"step": 63000
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 1.24,
|
| 782 |
+
"learning_rate": 1.9751395750019602e-05,
|
| 783 |
+
"loss": 4.0463,
|
| 784 |
+
"step": 63500
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 1.25,
|
| 788 |
+
"learning_rate": 1.974944326825061e-05,
|
| 789 |
+
"loss": 4.0845,
|
| 790 |
+
"step": 64000
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 1.26,
|
| 794 |
+
"learning_rate": 1.9747490786481614e-05,
|
| 795 |
+
"loss": 4.0752,
|
| 796 |
+
"step": 64500
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 1.27,
|
| 800 |
+
"learning_rate": 1.9745534384066496e-05,
|
| 801 |
+
"loss": 4.0262,
|
| 802 |
+
"step": 65000
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 1.28,
|
| 806 |
+
"learning_rate": 1.9743574061005255e-05,
|
| 807 |
+
"loss": 4.03,
|
| 808 |
+
"step": 65500
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.29,
|
| 812 |
+
"learning_rate": 1.9741617658590137e-05,
|
| 813 |
+
"loss": 4.0383,
|
| 814 |
+
"step": 66000
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.3,
|
| 818 |
+
"learning_rate": 1.9739657335528896e-05,
|
| 819 |
+
"loss": 4.0371,
|
| 820 |
+
"step": 66500
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 1.31,
|
| 824 |
+
"learning_rate": 1.9737697012467655e-05,
|
| 825 |
+
"loss": 4.0516,
|
| 826 |
+
"step": 67000
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 1.32,
|
| 830 |
+
"learning_rate": 1.9735740610052537e-05,
|
| 831 |
+
"loss": 4.029,
|
| 832 |
+
"step": 67500
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 1.33,
|
| 836 |
+
"learning_rate": 1.9733784207637422e-05,
|
| 837 |
+
"loss": 4.0567,
|
| 838 |
+
"step": 68000
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 1.34,
|
| 842 |
+
"learning_rate": 1.973182388457618e-05,
|
| 843 |
+
"loss": 4.0524,
|
| 844 |
+
"step": 68500
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 1.35,
|
| 848 |
+
"learning_rate": 1.972986356151494e-05,
|
| 849 |
+
"loss": 4.0215,
|
| 850 |
+
"step": 69000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.36,
|
| 854 |
+
"learning_rate": 1.97279032384537e-05,
|
| 855 |
+
"loss": 4.0614,
|
| 856 |
+
"step": 69500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.37,
|
| 860 |
+
"learning_rate": 1.972594291539246e-05,
|
| 861 |
+
"loss": 4.0398,
|
| 862 |
+
"step": 70000
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 1.38,
|
| 866 |
+
"learning_rate": 1.972398651297734e-05,
|
| 867 |
+
"loss": 4.0334,
|
| 868 |
+
"step": 70500
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 1.39,
|
| 872 |
+
"learning_rate": 1.97220261899161e-05,
|
| 873 |
+
"loss": 4.0458,
|
| 874 |
+
"step": 71000
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 1.4,
|
| 878 |
+
"learning_rate": 1.972006586685486e-05,
|
| 879 |
+
"loss": 4.051,
|
| 880 |
+
"step": 71500
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 1.41,
|
| 884 |
+
"learning_rate": 1.971810554379362e-05,
|
| 885 |
+
"loss": 4.0592,
|
| 886 |
+
"step": 72000
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 1.42,
|
| 890 |
+
"learning_rate": 1.97161491413785e-05,
|
| 891 |
+
"loss": 4.0235,
|
| 892 |
+
"step": 72500
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.43,
|
| 896 |
+
"learning_rate": 1.9714188818317262e-05,
|
| 897 |
+
"loss": 4.0349,
|
| 898 |
+
"step": 73000
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.44,
|
| 902 |
+
"learning_rate": 1.971223241590214e-05,
|
| 903 |
+
"loss": 4.066,
|
| 904 |
+
"step": 73500
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 1.45,
|
| 908 |
+
"learning_rate": 1.9710276013487025e-05,
|
| 909 |
+
"loss": 4.0533,
|
| 910 |
+
"step": 74000
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 1.46,
|
| 914 |
+
"learning_rate": 1.9708315690425785e-05,
|
| 915 |
+
"loss": 4.0526,
|
| 916 |
+
"step": 74500
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 1.47,
|
| 920 |
+
"learning_rate": 1.9706355367364544e-05,
|
| 921 |
+
"loss": 4.0422,
|
| 922 |
+
"step": 75000
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 1.48,
|
| 926 |
+
"learning_rate": 1.9704395044303303e-05,
|
| 927 |
+
"loss": 4.0314,
|
| 928 |
+
"step": 75500
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 1.49,
|
| 932 |
+
"learning_rate": 1.9702434721242062e-05,
|
| 933 |
+
"loss": 4.033,
|
| 934 |
+
"step": 76000
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.5,
|
| 938 |
+
"learning_rate": 1.970047439818082e-05,
|
| 939 |
+
"loss": 4.0314,
|
| 940 |
+
"step": 76500
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.51,
|
| 944 |
+
"learning_rate": 1.9698517995765703e-05,
|
| 945 |
+
"loss": 4.0606,
|
| 946 |
+
"step": 77000
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 1.52,
|
| 950 |
+
"learning_rate": 1.9696557672704462e-05,
|
| 951 |
+
"loss": 4.0381,
|
| 952 |
+
"step": 77500
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 1.53,
|
| 956 |
+
"learning_rate": 1.969459734964322e-05,
|
| 957 |
+
"loss": 4.0499,
|
| 958 |
+
"step": 78000
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 1.54,
|
| 962 |
+
"learning_rate": 1.969263702658198e-05,
|
| 963 |
+
"loss": 4.0226,
|
| 964 |
+
"step": 78500
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 1.55,
|
| 968 |
+
"learning_rate": 1.969067670352074e-05,
|
| 969 |
+
"loss": 4.0831,
|
| 970 |
+
"step": 79000
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 1.56,
|
| 974 |
+
"learning_rate": 1.96887163804595e-05,
|
| 975 |
+
"loss": 4.0174,
|
| 976 |
+
"step": 79500
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.57,
|
| 980 |
+
"learning_rate": 1.968675605739826e-05,
|
| 981 |
+
"loss": 4.026,
|
| 982 |
+
"step": 80000
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.58,
|
| 986 |
+
"learning_rate": 1.9684803575629266e-05,
|
| 987 |
+
"loss": 4.0229,
|
| 988 |
+
"step": 80500
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 1.59,
|
| 992 |
+
"learning_rate": 1.9682843252568025e-05,
|
| 993 |
+
"loss": 4.0314,
|
| 994 |
+
"step": 81000
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 1.6,
|
| 998 |
+
"learning_rate": 1.9680886850152907e-05,
|
| 999 |
+
"loss": 4.0725,
|
| 1000 |
+
"step": 81500
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 1.61,
|
| 1004 |
+
"learning_rate": 1.967893044773779e-05,
|
| 1005 |
+
"loss": 4.0628,
|
| 1006 |
+
"step": 82000
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 1.62,
|
| 1010 |
+
"learning_rate": 1.9676970124676548e-05,
|
| 1011 |
+
"loss": 4.0304,
|
| 1012 |
+
"step": 82500
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 1.63,
|
| 1016 |
+
"learning_rate": 1.9675009801615307e-05,
|
| 1017 |
+
"loss": 4.0507,
|
| 1018 |
+
"step": 83000
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.64,
|
| 1022 |
+
"learning_rate": 1.9673049478554066e-05,
|
| 1023 |
+
"loss": 4.0498,
|
| 1024 |
+
"step": 83500
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.65,
|
| 1028 |
+
"learning_rate": 1.9671089155492825e-05,
|
| 1029 |
+
"loss": 4.0216,
|
| 1030 |
+
"step": 84000
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 1.66,
|
| 1034 |
+
"learning_rate": 1.9669128832431584e-05,
|
| 1035 |
+
"loss": 4.0271,
|
| 1036 |
+
"step": 84500
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 1.67,
|
| 1040 |
+
"learning_rate": 1.9667168509370347e-05,
|
| 1041 |
+
"loss": 4.0393,
|
| 1042 |
+
"step": 85000
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 1.68,
|
| 1046 |
+
"learning_rate": 1.9665208186309106e-05,
|
| 1047 |
+
"loss": 4.0211,
|
| 1048 |
+
"step": 85500
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 1.69,
|
| 1052 |
+
"learning_rate": 1.9663247863247865e-05,
|
| 1053 |
+
"loss": 4.037,
|
| 1054 |
+
"step": 86000
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 1.7,
|
| 1058 |
+
"learning_rate": 1.9661287540186624e-05,
|
| 1059 |
+
"loss": 4.0525,
|
| 1060 |
+
"step": 86500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.71,
|
| 1064 |
+
"learning_rate": 1.9659331137771506e-05,
|
| 1065 |
+
"loss": 3.995,
|
| 1066 |
+
"step": 87000
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.72,
|
| 1070 |
+
"learning_rate": 1.965737865600251e-05,
|
| 1071 |
+
"loss": 4.0258,
|
| 1072 |
+
"step": 87500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 1.73,
|
| 1076 |
+
"learning_rate": 1.965541833294127e-05,
|
| 1077 |
+
"loss": 4.0427,
|
| 1078 |
+
"step": 88000
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 1.73,
|
| 1082 |
+
"learning_rate": 1.965346193052615e-05,
|
| 1083 |
+
"loss": 4.0075,
|
| 1084 |
+
"step": 88500
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 1.74,
|
| 1088 |
+
"learning_rate": 1.965150160746491e-05,
|
| 1089 |
+
"loss": 4.0431,
|
| 1090 |
+
"step": 89000
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 1.75,
|
| 1094 |
+
"learning_rate": 1.9649545205049792e-05,
|
| 1095 |
+
"loss": 4.0182,
|
| 1096 |
+
"step": 89500
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 1.76,
|
| 1100 |
+
"learning_rate": 1.964758488198855e-05,
|
| 1101 |
+
"loss": 4.0274,
|
| 1102 |
+
"step": 90000
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.77,
|
| 1106 |
+
"learning_rate": 1.9645628479573436e-05,
|
| 1107 |
+
"loss": 4.0351,
|
| 1108 |
+
"step": 90500
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 1.78,
|
| 1112 |
+
"learning_rate": 1.9643668156512195e-05,
|
| 1113 |
+
"loss": 4.0056,
|
| 1114 |
+
"step": 91000
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 1.79,
|
| 1118 |
+
"learning_rate": 1.9641707833450955e-05,
|
| 1119 |
+
"loss": 4.0573,
|
| 1120 |
+
"step": 91500
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 1.8,
|
| 1124 |
+
"learning_rate": 1.9639747510389714e-05,
|
| 1125 |
+
"loss": 4.0159,
|
| 1126 |
+
"step": 92000
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 1.81,
|
| 1130 |
+
"learning_rate": 1.9637791107974596e-05,
|
| 1131 |
+
"loss": 4.0638,
|
| 1132 |
+
"step": 92500
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 1.82,
|
| 1136 |
+
"learning_rate": 1.9635830784913355e-05,
|
| 1137 |
+
"loss": 4.026,
|
| 1138 |
+
"step": 93000
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 1.83,
|
| 1142 |
+
"learning_rate": 1.963387830314436e-05,
|
| 1143 |
+
"loss": 4.0216,
|
| 1144 |
+
"step": 93500
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.84,
|
| 1148 |
+
"learning_rate": 1.9631917980083118e-05,
|
| 1149 |
+
"loss": 4.0316,
|
| 1150 |
+
"step": 94000
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 1.85,
|
| 1154 |
+
"learning_rate": 1.9629957657021877e-05,
|
| 1155 |
+
"loss": 4.0415,
|
| 1156 |
+
"step": 94500
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 1.86,
|
| 1160 |
+
"learning_rate": 1.9627997333960636e-05,
|
| 1161 |
+
"loss": 4.0422,
|
| 1162 |
+
"step": 95000
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 1.87,
|
| 1166 |
+
"learning_rate": 1.9626037010899396e-05,
|
| 1167 |
+
"loss": 4.0336,
|
| 1168 |
+
"step": 95500
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 1.88,
|
| 1172 |
+
"learning_rate": 1.9624076687838158e-05,
|
| 1173 |
+
"loss": 4.0264,
|
| 1174 |
+
"step": 96000
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 1.89,
|
| 1178 |
+
"learning_rate": 1.9622116364776917e-05,
|
| 1179 |
+
"loss": 4.0486,
|
| 1180 |
+
"step": 96500
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 1.9,
|
| 1184 |
+
"learning_rate": 1.96201599623618e-05,
|
| 1185 |
+
"loss": 4.0507,
|
| 1186 |
+
"step": 97000
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.91,
|
| 1190 |
+
"learning_rate": 1.961819963930056e-05,
|
| 1191 |
+
"loss": 4.0275,
|
| 1192 |
+
"step": 97500
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 1.92,
|
| 1196 |
+
"learning_rate": 1.9616239316239318e-05,
|
| 1197 |
+
"loss": 4.0237,
|
| 1198 |
+
"step": 98000
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 1.93,
|
| 1202 |
+
"learning_rate": 1.9614278993178077e-05,
|
| 1203 |
+
"loss": 4.0183,
|
| 1204 |
+
"step": 98500
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 1.94,
|
| 1208 |
+
"learning_rate": 1.9612318670116836e-05,
|
| 1209 |
+
"loss": 4.0259,
|
| 1210 |
+
"step": 99000
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 1.95,
|
| 1214 |
+
"learning_rate": 1.9610358347055595e-05,
|
| 1215 |
+
"loss": 4.0354,
|
| 1216 |
+
"step": 99500
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 1.96,
|
| 1220 |
+
"learning_rate": 1.9608398023994354e-05,
|
| 1221 |
+
"loss": 4.0269,
|
| 1222 |
+
"step": 100000
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 1.97,
|
| 1226 |
+
"learning_rate": 1.9606441621579236e-05,
|
| 1227 |
+
"loss": 4.0241,
|
| 1228 |
+
"step": 100500
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.98,
|
| 1232 |
+
"learning_rate": 1.9604481298517995e-05,
|
| 1233 |
+
"loss": 4.0237,
|
| 1234 |
+
"step": 101000
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 1.99,
|
| 1238 |
+
"learning_rate": 1.9602520975456754e-05,
|
| 1239 |
+
"loss": 4.0526,
|
| 1240 |
+
"step": 101500
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 2.0,
|
| 1244 |
+
"learning_rate": 1.9600560652395517e-05,
|
| 1245 |
+
"loss": 4.0422,
|
| 1246 |
+
"step": 102000
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 2.0,
|
| 1250 |
+
"eval_gen_len": 18.9837,
|
| 1251 |
+
"eval_loss": 3.3488144874572754,
|
| 1252 |
+
"eval_rouge1": 0.1877,
|
| 1253 |
+
"eval_rouge2": 0.0439,
|
| 1254 |
+
"eval_rougeL": 0.1542,
|
| 1255 |
+
"eval_rougeLsum": 0.1541,
|
| 1256 |
+
"eval_runtime": 1358.2103,
|
| 1257 |
+
"eval_samples_per_second": 8.343,
|
| 1258 |
+
"eval_steps_per_second": 2.086,
|
| 1259 |
+
"step": 102024
|
| 1260 |
+
}
|
| 1261 |
+
],
|
| 1262 |
+
"logging_steps": 500,
|
| 1263 |
+
"max_steps": 5101200,
|
| 1264 |
+
"num_train_epochs": 100,
|
| 1265 |
+
"save_steps": 500,
|
| 1266 |
+
"total_flos": 3.920135247069235e+17,
|
| 1267 |
+
"trial_name": null,
|
| 1268 |
+
"trial_params": null
|
| 1269 |
+
}
|
checkpoint-102024/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1e39fc9a9199e5870ea9a20d0b50a05a4baebf235ae7dc6733b399c85396bb0e
|
| 3 |
+
size 4219
|