Training checkpoint 600
Browse files- checkpoint-400/added_tokens.json +1 -0
- checkpoint-400/config.json +60 -0
- checkpoint-400/optimizer.pt +3 -0
- checkpoint-400/pytorch_model.bin +3 -0
- checkpoint-400/rng_state.pth +3 -0
- checkpoint-400/scheduler.pt +3 -0
- checkpoint-400/special_tokens_map.json +1 -0
- checkpoint-400/spiece.model +3 -0
- checkpoint-400/tokenizer.json +0 -0
- checkpoint-400/tokenizer_config.json +1 -0
- checkpoint-400/trainer_state.json +640 -0
- checkpoint-400/training_args.bin +3 -0
- checkpoint-600/added_tokens.json +1 -0
- checkpoint-600/config.json +60 -0
- checkpoint-600/optimizer.pt +3 -0
- checkpoint-600/pytorch_model.bin +3 -0
- checkpoint-600/rng_state.pth +3 -0
- checkpoint-600/scheduler.pt +3 -0
- checkpoint-600/special_tokens_map.json +1 -0
- checkpoint-600/spiece.model +3 -0
- checkpoint-600/tokenizer.json +0 -0
- checkpoint-600/tokenizer_config.json +1 -0
- checkpoint-600/trainer_state.json +949 -0
- checkpoint-600/training_args.bin +3 -0
checkpoint-400/added_tokens.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{" <=": 32100, " <": 32101}
|
checkpoint-400/config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"diversity_penalty": null,
|
| 11 |
+
"dropout_rate": 0.1,
|
| 12 |
+
"eos_token_id": 1,
|
| 13 |
+
"feed_forward_proj": "relu",
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"is_encoder_decoder": true,
|
| 16 |
+
"layer_norm_epsilon": 1e-06,
|
| 17 |
+
"max_length": 512,
|
| 18 |
+
"model_type": "t5",
|
| 19 |
+
"n_positions": 512,
|
| 20 |
+
"num_decoder_layers": 6,
|
| 21 |
+
"num_heads": 8,
|
| 22 |
+
"num_layers": 6,
|
| 23 |
+
"num_relations": 48,
|
| 24 |
+
"output_past": true,
|
| 25 |
+
"pad_token_id": 0,
|
| 26 |
+
"relative_attention_num_buckets": 32,
|
| 27 |
+
"task_specific_params": {
|
| 28 |
+
"summarization": {
|
| 29 |
+
"early_stopping": true,
|
| 30 |
+
"length_penalty": 2.0,
|
| 31 |
+
"max_length": 200,
|
| 32 |
+
"min_length": 30,
|
| 33 |
+
"no_repeat_ngram_size": 3,
|
| 34 |
+
"num_beams": 4,
|
| 35 |
+
"prefix": "summarize: "
|
| 36 |
+
},
|
| 37 |
+
"translation_en_to_de": {
|
| 38 |
+
"early_stopping": true,
|
| 39 |
+
"max_length": 300,
|
| 40 |
+
"num_beams": 4,
|
| 41 |
+
"prefix": "translate English to German: "
|
| 42 |
+
},
|
| 43 |
+
"translation_en_to_fr": {
|
| 44 |
+
"early_stopping": true,
|
| 45 |
+
"max_length": 300,
|
| 46 |
+
"num_beams": 4,
|
| 47 |
+
"prefix": "translate English to French: "
|
| 48 |
+
},
|
| 49 |
+
"translation_en_to_ro": {
|
| 50 |
+
"early_stopping": true,
|
| 51 |
+
"max_length": 300,
|
| 52 |
+
"num_beams": 4,
|
| 53 |
+
"prefix": "translate English to Romanian: "
|
| 54 |
+
}
|
| 55 |
+
},
|
| 56 |
+
"torch_dtype": "float32",
|
| 57 |
+
"transformers_version": "4.11.3",
|
| 58 |
+
"use_cache": true,
|
| 59 |
+
"vocab_size": 32128
|
| 60 |
+
}
|
checkpoint-400/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9cd705d1b9f8a07973acfedf1c8294ccff14d6a83e69341bf856f9caea0ba6ea
|
| 3 |
+
size 840419
|
checkpoint-400/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:01d8f07bc06916b5b94752332fc9eb1c6f2b182ddb144101b80868347155b549
|
| 3 |
+
size 242111735
|
checkpoint-400/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b3cdec9cf5845ed6bbd01fa200c385f17770b411093edc31cdbd9494c7a11fc
|
| 3 |
+
size 14503
|
checkpoint-400/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64a7e411cfeebed3217cf4ae85e2b8a842c0a53b34d002639b1f5b126fa27288
|
| 3 |
+
size 623
|
checkpoint-400/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>"]}
|
checkpoint-400/spiece.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
|
| 3 |
+
size 791656
|
checkpoint-400/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-400/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", "tokenizer_class": "T5Tokenizer"}
|
checkpoint-400/trainer_state.json
ADDED
|
@@ -0,0 +1,640 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.48065764023210833,
|
| 3 |
+
"best_model_checkpoint": "train/DocuT5-RASAT-Small-SD/checkpoint-400",
|
| 4 |
+
"epoch": 399.93607305936075,
|
| 5 |
+
"global_step": 400,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.94,
|
| 12 |
+
"learning_rate": 0.001,
|
| 13 |
+
"loss": 4.452,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 3.94,
|
| 18 |
+
"learning_rate": 0.001,
|
| 19 |
+
"loss": 3.2202,
|
| 20 |
+
"step": 4
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 7.94,
|
| 24 |
+
"learning_rate": 0.001,
|
| 25 |
+
"loss": 2.0091,
|
| 26 |
+
"step": 8
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 11.94,
|
| 30 |
+
"learning_rate": 0.001,
|
| 31 |
+
"loss": 1.4161,
|
| 32 |
+
"step": 12
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 15.94,
|
| 36 |
+
"learning_rate": 0.001,
|
| 37 |
+
"loss": 1.055,
|
| 38 |
+
"step": 16
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 19.94,
|
| 42 |
+
"learning_rate": 0.001,
|
| 43 |
+
"loss": 0.9194,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 23.94,
|
| 48 |
+
"learning_rate": 0.001,
|
| 49 |
+
"loss": 0.6759,
|
| 50 |
+
"step": 24
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 27.94,
|
| 54 |
+
"learning_rate": 0.001,
|
| 55 |
+
"loss": 0.6637,
|
| 56 |
+
"step": 28
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 31.94,
|
| 60 |
+
"learning_rate": 0.001,
|
| 61 |
+
"loss": 0.4995,
|
| 62 |
+
"step": 32
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 35.94,
|
| 66 |
+
"learning_rate": 0.001,
|
| 67 |
+
"loss": 0.505,
|
| 68 |
+
"step": 36
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 39.94,
|
| 72 |
+
"learning_rate": 0.001,
|
| 73 |
+
"loss": 0.3915,
|
| 74 |
+
"step": 40
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 43.94,
|
| 78 |
+
"learning_rate": 0.001,
|
| 79 |
+
"loss": 0.3916,
|
| 80 |
+
"step": 44
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 47.94,
|
| 84 |
+
"learning_rate": 0.001,
|
| 85 |
+
"loss": 0.3585,
|
| 86 |
+
"step": 48
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 51.94,
|
| 90 |
+
"learning_rate": 0.001,
|
| 91 |
+
"loss": 0.3014,
|
| 92 |
+
"step": 52
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 55.94,
|
| 96 |
+
"learning_rate": 0.001,
|
| 97 |
+
"loss": 0.2761,
|
| 98 |
+
"step": 56
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 59.94,
|
| 102 |
+
"learning_rate": 0.001,
|
| 103 |
+
"loss": 0.2745,
|
| 104 |
+
"step": 60
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 63.94,
|
| 108 |
+
"learning_rate": 0.001,
|
| 109 |
+
"loss": 0.3122,
|
| 110 |
+
"step": 64
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 67.94,
|
| 114 |
+
"learning_rate": 0.001,
|
| 115 |
+
"loss": 0.234,
|
| 116 |
+
"step": 68
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 71.94,
|
| 120 |
+
"learning_rate": 0.001,
|
| 121 |
+
"loss": 0.2171,
|
| 122 |
+
"step": 72
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 75.94,
|
| 126 |
+
"learning_rate": 0.001,
|
| 127 |
+
"loss": 0.2122,
|
| 128 |
+
"step": 76
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 79.94,
|
| 132 |
+
"learning_rate": 0.001,
|
| 133 |
+
"loss": 0.1883,
|
| 134 |
+
"step": 80
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 83.94,
|
| 138 |
+
"learning_rate": 0.001,
|
| 139 |
+
"loss": 0.1895,
|
| 140 |
+
"step": 84
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 87.94,
|
| 144 |
+
"learning_rate": 0.001,
|
| 145 |
+
"loss": 0.1789,
|
| 146 |
+
"step": 88
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 91.94,
|
| 150 |
+
"learning_rate": 0.001,
|
| 151 |
+
"loss": 0.1731,
|
| 152 |
+
"step": 92
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 95.94,
|
| 156 |
+
"learning_rate": 0.001,
|
| 157 |
+
"loss": 0.1587,
|
| 158 |
+
"step": 96
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 99.94,
|
| 162 |
+
"learning_rate": 0.001,
|
| 163 |
+
"loss": 0.1459,
|
| 164 |
+
"step": 100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 103.94,
|
| 168 |
+
"learning_rate": 0.001,
|
| 169 |
+
"loss": 0.1501,
|
| 170 |
+
"step": 104
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 107.94,
|
| 174 |
+
"learning_rate": 0.001,
|
| 175 |
+
"loss": 0.1412,
|
| 176 |
+
"step": 108
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 111.94,
|
| 180 |
+
"learning_rate": 0.001,
|
| 181 |
+
"loss": 0.1292,
|
| 182 |
+
"step": 112
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 115.94,
|
| 186 |
+
"learning_rate": 0.001,
|
| 187 |
+
"loss": 0.1141,
|
| 188 |
+
"step": 116
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 119.94,
|
| 192 |
+
"learning_rate": 0.001,
|
| 193 |
+
"loss": 0.1193,
|
| 194 |
+
"step": 120
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 123.94,
|
| 198 |
+
"learning_rate": 0.001,
|
| 199 |
+
"loss": 0.1313,
|
| 200 |
+
"step": 124
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 127.94,
|
| 204 |
+
"learning_rate": 0.001,
|
| 205 |
+
"loss": 0.1131,
|
| 206 |
+
"step": 128
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 131.94,
|
| 210 |
+
"learning_rate": 0.001,
|
| 211 |
+
"loss": 0.1038,
|
| 212 |
+
"step": 132
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 135.94,
|
| 216 |
+
"learning_rate": 0.001,
|
| 217 |
+
"loss": 0.0951,
|
| 218 |
+
"step": 136
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 139.94,
|
| 222 |
+
"learning_rate": 0.001,
|
| 223 |
+
"loss": 0.0891,
|
| 224 |
+
"step": 140
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 143.94,
|
| 228 |
+
"learning_rate": 0.001,
|
| 229 |
+
"loss": 0.0978,
|
| 230 |
+
"step": 144
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 147.94,
|
| 234 |
+
"learning_rate": 0.001,
|
| 235 |
+
"loss": 0.1093,
|
| 236 |
+
"step": 148
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 151.94,
|
| 240 |
+
"learning_rate": 0.001,
|
| 241 |
+
"loss": 0.1101,
|
| 242 |
+
"step": 152
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 155.94,
|
| 246 |
+
"learning_rate": 0.001,
|
| 247 |
+
"loss": 0.0945,
|
| 248 |
+
"step": 156
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 159.94,
|
| 252 |
+
"learning_rate": 0.001,
|
| 253 |
+
"loss": 0.0764,
|
| 254 |
+
"step": 160
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 163.94,
|
| 258 |
+
"learning_rate": 0.001,
|
| 259 |
+
"loss": 0.0884,
|
| 260 |
+
"step": 164
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 167.94,
|
| 264 |
+
"learning_rate": 0.001,
|
| 265 |
+
"loss": 0.1142,
|
| 266 |
+
"step": 168
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 171.94,
|
| 270 |
+
"learning_rate": 0.001,
|
| 271 |
+
"loss": 0.1011,
|
| 272 |
+
"step": 172
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 175.94,
|
| 276 |
+
"learning_rate": 0.001,
|
| 277 |
+
"loss": 0.0702,
|
| 278 |
+
"step": 176
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 179.94,
|
| 282 |
+
"learning_rate": 0.001,
|
| 283 |
+
"loss": 0.0597,
|
| 284 |
+
"step": 180
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 183.94,
|
| 288 |
+
"learning_rate": 0.001,
|
| 289 |
+
"loss": 0.0579,
|
| 290 |
+
"step": 184
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 187.94,
|
| 294 |
+
"learning_rate": 0.001,
|
| 295 |
+
"loss": 0.0559,
|
| 296 |
+
"step": 188
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 191.94,
|
| 300 |
+
"learning_rate": 0.001,
|
| 301 |
+
"loss": 0.0558,
|
| 302 |
+
"step": 192
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 195.94,
|
| 306 |
+
"learning_rate": 0.001,
|
| 307 |
+
"loss": 0.0534,
|
| 308 |
+
"step": 196
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 199.94,
|
| 312 |
+
"learning_rate": 0.001,
|
| 313 |
+
"loss": 0.0505,
|
| 314 |
+
"step": 200
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 199.94,
|
| 318 |
+
"eval_exact_match": 0.21083172147001933,
|
| 319 |
+
"eval_exec": 0.23017408123791103,
|
| 320 |
+
"eval_loss": 0.37137603759765625,
|
| 321 |
+
"eval_runtime": 103.3205,
|
| 322 |
+
"eval_samples_per_second": 10.008,
|
| 323 |
+
"step": 200
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 203.94,
|
| 327 |
+
"learning_rate": 0.001,
|
| 328 |
+
"loss": 0.0466,
|
| 329 |
+
"step": 204
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 207.94,
|
| 333 |
+
"learning_rate": 0.001,
|
| 334 |
+
"loss": 0.0472,
|
| 335 |
+
"step": 208
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 211.94,
|
| 339 |
+
"learning_rate": 0.001,
|
| 340 |
+
"loss": 0.0474,
|
| 341 |
+
"step": 212
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 215.94,
|
| 345 |
+
"learning_rate": 0.001,
|
| 346 |
+
"loss": 0.0439,
|
| 347 |
+
"step": 216
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 219.94,
|
| 351 |
+
"learning_rate": 0.001,
|
| 352 |
+
"loss": 0.0536,
|
| 353 |
+
"step": 220
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 223.94,
|
| 357 |
+
"learning_rate": 0.001,
|
| 358 |
+
"loss": 0.0673,
|
| 359 |
+
"step": 224
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 227.94,
|
| 363 |
+
"learning_rate": 0.001,
|
| 364 |
+
"loss": 0.0939,
|
| 365 |
+
"step": 228
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 231.94,
|
| 369 |
+
"learning_rate": 0.001,
|
| 370 |
+
"loss": 0.0489,
|
| 371 |
+
"step": 232
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 235.94,
|
| 375 |
+
"learning_rate": 0.001,
|
| 376 |
+
"loss": 0.0363,
|
| 377 |
+
"step": 236
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 239.94,
|
| 381 |
+
"learning_rate": 0.001,
|
| 382 |
+
"loss": 0.0375,
|
| 383 |
+
"step": 240
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 243.94,
|
| 387 |
+
"learning_rate": 0.001,
|
| 388 |
+
"loss": 0.0351,
|
| 389 |
+
"step": 244
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 247.94,
|
| 393 |
+
"learning_rate": 0.001,
|
| 394 |
+
"loss": 0.0315,
|
| 395 |
+
"step": 248
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 251.94,
|
| 399 |
+
"learning_rate": 0.001,
|
| 400 |
+
"loss": 0.0303,
|
| 401 |
+
"step": 252
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 255.94,
|
| 405 |
+
"learning_rate": 0.001,
|
| 406 |
+
"loss": 0.041,
|
| 407 |
+
"step": 256
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 259.94,
|
| 411 |
+
"learning_rate": 0.001,
|
| 412 |
+
"loss": 0.0489,
|
| 413 |
+
"step": 260
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 263.94,
|
| 417 |
+
"learning_rate": 0.001,
|
| 418 |
+
"loss": 0.0536,
|
| 419 |
+
"step": 264
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 267.94,
|
| 423 |
+
"learning_rate": 0.001,
|
| 424 |
+
"loss": 0.0338,
|
| 425 |
+
"step": 268
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 271.94,
|
| 429 |
+
"learning_rate": 0.001,
|
| 430 |
+
"loss": 0.0266,
|
| 431 |
+
"step": 272
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 275.94,
|
| 435 |
+
"learning_rate": 0.001,
|
| 436 |
+
"loss": 0.0259,
|
| 437 |
+
"step": 276
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 279.94,
|
| 441 |
+
"learning_rate": 0.001,
|
| 442 |
+
"loss": 0.0238,
|
| 443 |
+
"step": 280
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 283.94,
|
| 447 |
+
"learning_rate": 0.001,
|
| 448 |
+
"loss": 0.0235,
|
| 449 |
+
"step": 284
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 287.94,
|
| 453 |
+
"learning_rate": 0.001,
|
| 454 |
+
"loss": 0.0226,
|
| 455 |
+
"step": 288
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 291.94,
|
| 459 |
+
"learning_rate": 0.001,
|
| 460 |
+
"loss": 0.0213,
|
| 461 |
+
"step": 292
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 295.94,
|
| 465 |
+
"learning_rate": 0.001,
|
| 466 |
+
"loss": 0.0222,
|
| 467 |
+
"step": 296
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 299.94,
|
| 471 |
+
"learning_rate": 0.001,
|
| 472 |
+
"loss": 0.0226,
|
| 473 |
+
"step": 300
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 303.94,
|
| 477 |
+
"learning_rate": 0.001,
|
| 478 |
+
"loss": 0.0262,
|
| 479 |
+
"step": 304
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 307.94,
|
| 483 |
+
"learning_rate": 0.001,
|
| 484 |
+
"loss": 0.0251,
|
| 485 |
+
"step": 308
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 311.94,
|
| 489 |
+
"learning_rate": 0.001,
|
| 490 |
+
"loss": 0.0184,
|
| 491 |
+
"step": 312
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 315.94,
|
| 495 |
+
"learning_rate": 0.001,
|
| 496 |
+
"loss": 0.0209,
|
| 497 |
+
"step": 316
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 319.94,
|
| 501 |
+
"learning_rate": 0.001,
|
| 502 |
+
"loss": 0.0176,
|
| 503 |
+
"step": 320
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 323.94,
|
| 507 |
+
"learning_rate": 0.001,
|
| 508 |
+
"loss": 0.017,
|
| 509 |
+
"step": 324
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 327.94,
|
| 513 |
+
"learning_rate": 0.001,
|
| 514 |
+
"loss": 0.0163,
|
| 515 |
+
"step": 328
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 331.94,
|
| 519 |
+
"learning_rate": 0.001,
|
| 520 |
+
"loss": 0.0162,
|
| 521 |
+
"step": 332
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 335.94,
|
| 525 |
+
"learning_rate": 0.001,
|
| 526 |
+
"loss": 0.0188,
|
| 527 |
+
"step": 336
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 339.94,
|
| 531 |
+
"learning_rate": 0.001,
|
| 532 |
+
"loss": 0.0168,
|
| 533 |
+
"step": 340
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 343.94,
|
| 537 |
+
"learning_rate": 0.001,
|
| 538 |
+
"loss": 0.0139,
|
| 539 |
+
"step": 344
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 347.94,
|
| 543 |
+
"learning_rate": 0.001,
|
| 544 |
+
"loss": 0.0147,
|
| 545 |
+
"step": 348
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 351.94,
|
| 549 |
+
"learning_rate": 0.001,
|
| 550 |
+
"loss": 0.0144,
|
| 551 |
+
"step": 352
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 355.94,
|
| 555 |
+
"learning_rate": 0.001,
|
| 556 |
+
"loss": 0.0129,
|
| 557 |
+
"step": 356
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 359.94,
|
| 561 |
+
"learning_rate": 0.001,
|
| 562 |
+
"loss": 0.0214,
|
| 563 |
+
"step": 360
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 363.94,
|
| 567 |
+
"learning_rate": 0.001,
|
| 568 |
+
"loss": 0.0602,
|
| 569 |
+
"step": 364
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 367.94,
|
| 573 |
+
"learning_rate": 0.001,
|
| 574 |
+
"loss": 0.031,
|
| 575 |
+
"step": 368
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 371.94,
|
| 579 |
+
"learning_rate": 0.001,
|
| 580 |
+
"loss": 0.0167,
|
| 581 |
+
"step": 372
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 375.94,
|
| 585 |
+
"learning_rate": 0.001,
|
| 586 |
+
"loss": 0.0119,
|
| 587 |
+
"step": 376
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 379.94,
|
| 591 |
+
"learning_rate": 0.001,
|
| 592 |
+
"loss": 0.0115,
|
| 593 |
+
"step": 380
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 383.94,
|
| 597 |
+
"learning_rate": 0.001,
|
| 598 |
+
"loss": 0.0107,
|
| 599 |
+
"step": 384
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 387.94,
|
| 603 |
+
"learning_rate": 0.001,
|
| 604 |
+
"loss": 0.0099,
|
| 605 |
+
"step": 388
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 391.94,
|
| 609 |
+
"learning_rate": 0.001,
|
| 610 |
+
"loss": 0.0097,
|
| 611 |
+
"step": 392
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 395.94,
|
| 615 |
+
"learning_rate": 0.001,
|
| 616 |
+
"loss": 0.009,
|
| 617 |
+
"step": 396
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 399.94,
|
| 621 |
+
"learning_rate": 0.001,
|
| 622 |
+
"loss": 0.0086,
|
| 623 |
+
"step": 400
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 399.94,
|
| 627 |
+
"eval_exact_match": 0.48065764023210833,
|
| 628 |
+
"eval_exec": 0.4758220502901354,
|
| 629 |
+
"eval_loss": 0.47830089926719666,
|
| 630 |
+
"eval_runtime": 106.7292,
|
| 631 |
+
"eval_samples_per_second": 9.688,
|
| 632 |
+
"step": 400
|
| 633 |
+
}
|
| 634 |
+
],
|
| 635 |
+
"max_steps": 1602,
|
| 636 |
+
"num_train_epochs": 1602,
|
| 637 |
+
"total_flos": 4.576454547800064e+17,
|
| 638 |
+
"trial_name": null,
|
| 639 |
+
"trial_params": null
|
| 640 |
+
}
|
checkpoint-400/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c20e4fbfc3c6e458b39e5e8bb30e9452245ef3516a01db5a085e691f53119d55
|
| 3 |
+
size 3055
|
checkpoint-600/added_tokens.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{" <=": 32100, " <": 32101}
|
checkpoint-600/config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"diversity_penalty": null,
|
| 11 |
+
"dropout_rate": 0.1,
|
| 12 |
+
"eos_token_id": 1,
|
| 13 |
+
"feed_forward_proj": "relu",
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"is_encoder_decoder": true,
|
| 16 |
+
"layer_norm_epsilon": 1e-06,
|
| 17 |
+
"max_length": 512,
|
| 18 |
+
"model_type": "t5",
|
| 19 |
+
"n_positions": 512,
|
| 20 |
+
"num_decoder_layers": 6,
|
| 21 |
+
"num_heads": 8,
|
| 22 |
+
"num_layers": 6,
|
| 23 |
+
"num_relations": 48,
|
| 24 |
+
"output_past": true,
|
| 25 |
+
"pad_token_id": 0,
|
| 26 |
+
"relative_attention_num_buckets": 32,
|
| 27 |
+
"task_specific_params": {
|
| 28 |
+
"summarization": {
|
| 29 |
+
"early_stopping": true,
|
| 30 |
+
"length_penalty": 2.0,
|
| 31 |
+
"max_length": 200,
|
| 32 |
+
"min_length": 30,
|
| 33 |
+
"no_repeat_ngram_size": 3,
|
| 34 |
+
"num_beams": 4,
|
| 35 |
+
"prefix": "summarize: "
|
| 36 |
+
},
|
| 37 |
+
"translation_en_to_de": {
|
| 38 |
+
"early_stopping": true,
|
| 39 |
+
"max_length": 300,
|
| 40 |
+
"num_beams": 4,
|
| 41 |
+
"prefix": "translate English to German: "
|
| 42 |
+
},
|
| 43 |
+
"translation_en_to_fr": {
|
| 44 |
+
"early_stopping": true,
|
| 45 |
+
"max_length": 300,
|
| 46 |
+
"num_beams": 4,
|
| 47 |
+
"prefix": "translate English to French: "
|
| 48 |
+
},
|
| 49 |
+
"translation_en_to_ro": {
|
| 50 |
+
"early_stopping": true,
|
| 51 |
+
"max_length": 300,
|
| 52 |
+
"num_beams": 4,
|
| 53 |
+
"prefix": "translate English to Romanian: "
|
| 54 |
+
}
|
| 55 |
+
},
|
| 56 |
+
"torch_dtype": "float32",
|
| 57 |
+
"transformers_version": "4.11.3",
|
| 58 |
+
"use_cache": true,
|
| 59 |
+
"vocab_size": 32128
|
| 60 |
+
}
|
checkpoint-600/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cf9e5ef53cab4a00221a39f9131819dd744f7ede13eed86f171c94b2b5148595
|
| 3 |
+
size 840419
|
checkpoint-600/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e22ddbfee3e19f00dc475a1dea009563dc2bb593b509ba335eb04e82c6544e47
|
| 3 |
+
size 242111735
|
checkpoint-600/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9d508c0b69f1006fa2db090d150db78676d9ee95c99c85a5461a916654c8289
|
| 3 |
+
size 14503
|
checkpoint-600/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6b13200c02f19611a0f4f9972ffc5332b96e1112e0ce2e774a55fa8dc91a2be4
|
| 3 |
+
size 623
|
checkpoint-600/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>"]}
|
checkpoint-600/spiece.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
|
| 3 |
+
size 791656
|
checkpoint-600/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-600/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", "tokenizer_class": "T5Tokenizer"}
|
checkpoint-600/trainer_state.json
ADDED
|
@@ -0,0 +1,949 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.48065764023210833,
|
| 3 |
+
"best_model_checkpoint": "train/DocuT5-RASAT-Small-SD/checkpoint-400",
|
| 4 |
+
"epoch": 599.9360730593608,
|
| 5 |
+
"global_step": 600,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.94,
|
| 12 |
+
"learning_rate": 0.001,
|
| 13 |
+
"loss": 4.452,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 3.94,
|
| 18 |
+
"learning_rate": 0.001,
|
| 19 |
+
"loss": 3.2202,
|
| 20 |
+
"step": 4
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 7.94,
|
| 24 |
+
"learning_rate": 0.001,
|
| 25 |
+
"loss": 2.0091,
|
| 26 |
+
"step": 8
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 11.94,
|
| 30 |
+
"learning_rate": 0.001,
|
| 31 |
+
"loss": 1.4161,
|
| 32 |
+
"step": 12
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 15.94,
|
| 36 |
+
"learning_rate": 0.001,
|
| 37 |
+
"loss": 1.055,
|
| 38 |
+
"step": 16
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 19.94,
|
| 42 |
+
"learning_rate": 0.001,
|
| 43 |
+
"loss": 0.9194,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 23.94,
|
| 48 |
+
"learning_rate": 0.001,
|
| 49 |
+
"loss": 0.6759,
|
| 50 |
+
"step": 24
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 27.94,
|
| 54 |
+
"learning_rate": 0.001,
|
| 55 |
+
"loss": 0.6637,
|
| 56 |
+
"step": 28
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 31.94,
|
| 60 |
+
"learning_rate": 0.001,
|
| 61 |
+
"loss": 0.4995,
|
| 62 |
+
"step": 32
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 35.94,
|
| 66 |
+
"learning_rate": 0.001,
|
| 67 |
+
"loss": 0.505,
|
| 68 |
+
"step": 36
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 39.94,
|
| 72 |
+
"learning_rate": 0.001,
|
| 73 |
+
"loss": 0.3915,
|
| 74 |
+
"step": 40
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 43.94,
|
| 78 |
+
"learning_rate": 0.001,
|
| 79 |
+
"loss": 0.3916,
|
| 80 |
+
"step": 44
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 47.94,
|
| 84 |
+
"learning_rate": 0.001,
|
| 85 |
+
"loss": 0.3585,
|
| 86 |
+
"step": 48
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 51.94,
|
| 90 |
+
"learning_rate": 0.001,
|
| 91 |
+
"loss": 0.3014,
|
| 92 |
+
"step": 52
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 55.94,
|
| 96 |
+
"learning_rate": 0.001,
|
| 97 |
+
"loss": 0.2761,
|
| 98 |
+
"step": 56
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 59.94,
|
| 102 |
+
"learning_rate": 0.001,
|
| 103 |
+
"loss": 0.2745,
|
| 104 |
+
"step": 60
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 63.94,
|
| 108 |
+
"learning_rate": 0.001,
|
| 109 |
+
"loss": 0.3122,
|
| 110 |
+
"step": 64
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 67.94,
|
| 114 |
+
"learning_rate": 0.001,
|
| 115 |
+
"loss": 0.234,
|
| 116 |
+
"step": 68
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 71.94,
|
| 120 |
+
"learning_rate": 0.001,
|
| 121 |
+
"loss": 0.2171,
|
| 122 |
+
"step": 72
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 75.94,
|
| 126 |
+
"learning_rate": 0.001,
|
| 127 |
+
"loss": 0.2122,
|
| 128 |
+
"step": 76
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 79.94,
|
| 132 |
+
"learning_rate": 0.001,
|
| 133 |
+
"loss": 0.1883,
|
| 134 |
+
"step": 80
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 83.94,
|
| 138 |
+
"learning_rate": 0.001,
|
| 139 |
+
"loss": 0.1895,
|
| 140 |
+
"step": 84
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 87.94,
|
| 144 |
+
"learning_rate": 0.001,
|
| 145 |
+
"loss": 0.1789,
|
| 146 |
+
"step": 88
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 91.94,
|
| 150 |
+
"learning_rate": 0.001,
|
| 151 |
+
"loss": 0.1731,
|
| 152 |
+
"step": 92
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 95.94,
|
| 156 |
+
"learning_rate": 0.001,
|
| 157 |
+
"loss": 0.1587,
|
| 158 |
+
"step": 96
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 99.94,
|
| 162 |
+
"learning_rate": 0.001,
|
| 163 |
+
"loss": 0.1459,
|
| 164 |
+
"step": 100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 103.94,
|
| 168 |
+
"learning_rate": 0.001,
|
| 169 |
+
"loss": 0.1501,
|
| 170 |
+
"step": 104
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 107.94,
|
| 174 |
+
"learning_rate": 0.001,
|
| 175 |
+
"loss": 0.1412,
|
| 176 |
+
"step": 108
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 111.94,
|
| 180 |
+
"learning_rate": 0.001,
|
| 181 |
+
"loss": 0.1292,
|
| 182 |
+
"step": 112
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 115.94,
|
| 186 |
+
"learning_rate": 0.001,
|
| 187 |
+
"loss": 0.1141,
|
| 188 |
+
"step": 116
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 119.94,
|
| 192 |
+
"learning_rate": 0.001,
|
| 193 |
+
"loss": 0.1193,
|
| 194 |
+
"step": 120
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 123.94,
|
| 198 |
+
"learning_rate": 0.001,
|
| 199 |
+
"loss": 0.1313,
|
| 200 |
+
"step": 124
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 127.94,
|
| 204 |
+
"learning_rate": 0.001,
|
| 205 |
+
"loss": 0.1131,
|
| 206 |
+
"step": 128
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 131.94,
|
| 210 |
+
"learning_rate": 0.001,
|
| 211 |
+
"loss": 0.1038,
|
| 212 |
+
"step": 132
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 135.94,
|
| 216 |
+
"learning_rate": 0.001,
|
| 217 |
+
"loss": 0.0951,
|
| 218 |
+
"step": 136
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 139.94,
|
| 222 |
+
"learning_rate": 0.001,
|
| 223 |
+
"loss": 0.0891,
|
| 224 |
+
"step": 140
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 143.94,
|
| 228 |
+
"learning_rate": 0.001,
|
| 229 |
+
"loss": 0.0978,
|
| 230 |
+
"step": 144
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 147.94,
|
| 234 |
+
"learning_rate": 0.001,
|
| 235 |
+
"loss": 0.1093,
|
| 236 |
+
"step": 148
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 151.94,
|
| 240 |
+
"learning_rate": 0.001,
|
| 241 |
+
"loss": 0.1101,
|
| 242 |
+
"step": 152
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 155.94,
|
| 246 |
+
"learning_rate": 0.001,
|
| 247 |
+
"loss": 0.0945,
|
| 248 |
+
"step": 156
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 159.94,
|
| 252 |
+
"learning_rate": 0.001,
|
| 253 |
+
"loss": 0.0764,
|
| 254 |
+
"step": 160
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 163.94,
|
| 258 |
+
"learning_rate": 0.001,
|
| 259 |
+
"loss": 0.0884,
|
| 260 |
+
"step": 164
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 167.94,
|
| 264 |
+
"learning_rate": 0.001,
|
| 265 |
+
"loss": 0.1142,
|
| 266 |
+
"step": 168
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 171.94,
|
| 270 |
+
"learning_rate": 0.001,
|
| 271 |
+
"loss": 0.1011,
|
| 272 |
+
"step": 172
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 175.94,
|
| 276 |
+
"learning_rate": 0.001,
|
| 277 |
+
"loss": 0.0702,
|
| 278 |
+
"step": 176
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 179.94,
|
| 282 |
+
"learning_rate": 0.001,
|
| 283 |
+
"loss": 0.0597,
|
| 284 |
+
"step": 180
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 183.94,
|
| 288 |
+
"learning_rate": 0.001,
|
| 289 |
+
"loss": 0.0579,
|
| 290 |
+
"step": 184
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 187.94,
|
| 294 |
+
"learning_rate": 0.001,
|
| 295 |
+
"loss": 0.0559,
|
| 296 |
+
"step": 188
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 191.94,
|
| 300 |
+
"learning_rate": 0.001,
|
| 301 |
+
"loss": 0.0558,
|
| 302 |
+
"step": 192
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 195.94,
|
| 306 |
+
"learning_rate": 0.001,
|
| 307 |
+
"loss": 0.0534,
|
| 308 |
+
"step": 196
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 199.94,
|
| 312 |
+
"learning_rate": 0.001,
|
| 313 |
+
"loss": 0.0505,
|
| 314 |
+
"step": 200
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 199.94,
|
| 318 |
+
"eval_exact_match": 0.21083172147001933,
|
| 319 |
+
"eval_exec": 0.23017408123791103,
|
| 320 |
+
"eval_loss": 0.37137603759765625,
|
| 321 |
+
"eval_runtime": 103.3205,
|
| 322 |
+
"eval_samples_per_second": 10.008,
|
| 323 |
+
"step": 200
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 203.94,
|
| 327 |
+
"learning_rate": 0.001,
|
| 328 |
+
"loss": 0.0466,
|
| 329 |
+
"step": 204
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 207.94,
|
| 333 |
+
"learning_rate": 0.001,
|
| 334 |
+
"loss": 0.0472,
|
| 335 |
+
"step": 208
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 211.94,
|
| 339 |
+
"learning_rate": 0.001,
|
| 340 |
+
"loss": 0.0474,
|
| 341 |
+
"step": 212
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 215.94,
|
| 345 |
+
"learning_rate": 0.001,
|
| 346 |
+
"loss": 0.0439,
|
| 347 |
+
"step": 216
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 219.94,
|
| 351 |
+
"learning_rate": 0.001,
|
| 352 |
+
"loss": 0.0536,
|
| 353 |
+
"step": 220
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 223.94,
|
| 357 |
+
"learning_rate": 0.001,
|
| 358 |
+
"loss": 0.0673,
|
| 359 |
+
"step": 224
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 227.94,
|
| 363 |
+
"learning_rate": 0.001,
|
| 364 |
+
"loss": 0.0939,
|
| 365 |
+
"step": 228
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 231.94,
|
| 369 |
+
"learning_rate": 0.001,
|
| 370 |
+
"loss": 0.0489,
|
| 371 |
+
"step": 232
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 235.94,
|
| 375 |
+
"learning_rate": 0.001,
|
| 376 |
+
"loss": 0.0363,
|
| 377 |
+
"step": 236
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 239.94,
|
| 381 |
+
"learning_rate": 0.001,
|
| 382 |
+
"loss": 0.0375,
|
| 383 |
+
"step": 240
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 243.94,
|
| 387 |
+
"learning_rate": 0.001,
|
| 388 |
+
"loss": 0.0351,
|
| 389 |
+
"step": 244
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 247.94,
|
| 393 |
+
"learning_rate": 0.001,
|
| 394 |
+
"loss": 0.0315,
|
| 395 |
+
"step": 248
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 251.94,
|
| 399 |
+
"learning_rate": 0.001,
|
| 400 |
+
"loss": 0.0303,
|
| 401 |
+
"step": 252
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 255.94,
|
| 405 |
+
"learning_rate": 0.001,
|
| 406 |
+
"loss": 0.041,
|
| 407 |
+
"step": 256
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 259.94,
|
| 411 |
+
"learning_rate": 0.001,
|
| 412 |
+
"loss": 0.0489,
|
| 413 |
+
"step": 260
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 263.94,
|
| 417 |
+
"learning_rate": 0.001,
|
| 418 |
+
"loss": 0.0536,
|
| 419 |
+
"step": 264
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 267.94,
|
| 423 |
+
"learning_rate": 0.001,
|
| 424 |
+
"loss": 0.0338,
|
| 425 |
+
"step": 268
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 271.94,
|
| 429 |
+
"learning_rate": 0.001,
|
| 430 |
+
"loss": 0.0266,
|
| 431 |
+
"step": 272
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 275.94,
|
| 435 |
+
"learning_rate": 0.001,
|
| 436 |
+
"loss": 0.0259,
|
| 437 |
+
"step": 276
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 279.94,
|
| 441 |
+
"learning_rate": 0.001,
|
| 442 |
+
"loss": 0.0238,
|
| 443 |
+
"step": 280
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 283.94,
|
| 447 |
+
"learning_rate": 0.001,
|
| 448 |
+
"loss": 0.0235,
|
| 449 |
+
"step": 284
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 287.94,
|
| 453 |
+
"learning_rate": 0.001,
|
| 454 |
+
"loss": 0.0226,
|
| 455 |
+
"step": 288
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 291.94,
|
| 459 |
+
"learning_rate": 0.001,
|
| 460 |
+
"loss": 0.0213,
|
| 461 |
+
"step": 292
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 295.94,
|
| 465 |
+
"learning_rate": 0.001,
|
| 466 |
+
"loss": 0.0222,
|
| 467 |
+
"step": 296
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 299.94,
|
| 471 |
+
"learning_rate": 0.001,
|
| 472 |
+
"loss": 0.0226,
|
| 473 |
+
"step": 300
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 303.94,
|
| 477 |
+
"learning_rate": 0.001,
|
| 478 |
+
"loss": 0.0262,
|
| 479 |
+
"step": 304
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 307.94,
|
| 483 |
+
"learning_rate": 0.001,
|
| 484 |
+
"loss": 0.0251,
|
| 485 |
+
"step": 308
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 311.94,
|
| 489 |
+
"learning_rate": 0.001,
|
| 490 |
+
"loss": 0.0184,
|
| 491 |
+
"step": 312
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 315.94,
|
| 495 |
+
"learning_rate": 0.001,
|
| 496 |
+
"loss": 0.0209,
|
| 497 |
+
"step": 316
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 319.94,
|
| 501 |
+
"learning_rate": 0.001,
|
| 502 |
+
"loss": 0.0176,
|
| 503 |
+
"step": 320
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 323.94,
|
| 507 |
+
"learning_rate": 0.001,
|
| 508 |
+
"loss": 0.017,
|
| 509 |
+
"step": 324
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 327.94,
|
| 513 |
+
"learning_rate": 0.001,
|
| 514 |
+
"loss": 0.0163,
|
| 515 |
+
"step": 328
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 331.94,
|
| 519 |
+
"learning_rate": 0.001,
|
| 520 |
+
"loss": 0.0162,
|
| 521 |
+
"step": 332
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 335.94,
|
| 525 |
+
"learning_rate": 0.001,
|
| 526 |
+
"loss": 0.0188,
|
| 527 |
+
"step": 336
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 339.94,
|
| 531 |
+
"learning_rate": 0.001,
|
| 532 |
+
"loss": 0.0168,
|
| 533 |
+
"step": 340
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 343.94,
|
| 537 |
+
"learning_rate": 0.001,
|
| 538 |
+
"loss": 0.0139,
|
| 539 |
+
"step": 344
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 347.94,
|
| 543 |
+
"learning_rate": 0.001,
|
| 544 |
+
"loss": 0.0147,
|
| 545 |
+
"step": 348
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 351.94,
|
| 549 |
+
"learning_rate": 0.001,
|
| 550 |
+
"loss": 0.0144,
|
| 551 |
+
"step": 352
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 355.94,
|
| 555 |
+
"learning_rate": 0.001,
|
| 556 |
+
"loss": 0.0129,
|
| 557 |
+
"step": 356
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 359.94,
|
| 561 |
+
"learning_rate": 0.001,
|
| 562 |
+
"loss": 0.0214,
|
| 563 |
+
"step": 360
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 363.94,
|
| 567 |
+
"learning_rate": 0.001,
|
| 568 |
+
"loss": 0.0602,
|
| 569 |
+
"step": 364
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 367.94,
|
| 573 |
+
"learning_rate": 0.001,
|
| 574 |
+
"loss": 0.031,
|
| 575 |
+
"step": 368
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 371.94,
|
| 579 |
+
"learning_rate": 0.001,
|
| 580 |
+
"loss": 0.0167,
|
| 581 |
+
"step": 372
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 375.94,
|
| 585 |
+
"learning_rate": 0.001,
|
| 586 |
+
"loss": 0.0119,
|
| 587 |
+
"step": 376
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 379.94,
|
| 591 |
+
"learning_rate": 0.001,
|
| 592 |
+
"loss": 0.0115,
|
| 593 |
+
"step": 380
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 383.94,
|
| 597 |
+
"learning_rate": 0.001,
|
| 598 |
+
"loss": 0.0107,
|
| 599 |
+
"step": 384
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 387.94,
|
| 603 |
+
"learning_rate": 0.001,
|
| 604 |
+
"loss": 0.0099,
|
| 605 |
+
"step": 388
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 391.94,
|
| 609 |
+
"learning_rate": 0.001,
|
| 610 |
+
"loss": 0.0097,
|
| 611 |
+
"step": 392
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 395.94,
|
| 615 |
+
"learning_rate": 0.001,
|
| 616 |
+
"loss": 0.009,
|
| 617 |
+
"step": 396
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 399.94,
|
| 621 |
+
"learning_rate": 0.001,
|
| 622 |
+
"loss": 0.0086,
|
| 623 |
+
"step": 400
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 399.94,
|
| 627 |
+
"eval_exact_match": 0.48065764023210833,
|
| 628 |
+
"eval_exec": 0.4758220502901354,
|
| 629 |
+
"eval_loss": 0.47830089926719666,
|
| 630 |
+
"eval_runtime": 106.7292,
|
| 631 |
+
"eval_samples_per_second": 9.688,
|
| 632 |
+
"step": 400
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 403.94,
|
| 636 |
+
"learning_rate": 0.001,
|
| 637 |
+
"loss": 0.0085,
|
| 638 |
+
"step": 404
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 407.94,
|
| 642 |
+
"learning_rate": 0.001,
|
| 643 |
+
"loss": 0.0087,
|
| 644 |
+
"step": 408
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 411.94,
|
| 648 |
+
"learning_rate": 0.001,
|
| 649 |
+
"loss": 0.0089,
|
| 650 |
+
"step": 412
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 415.94,
|
| 654 |
+
"learning_rate": 0.001,
|
| 655 |
+
"loss": 0.0079,
|
| 656 |
+
"step": 416
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 419.94,
|
| 660 |
+
"learning_rate": 0.001,
|
| 661 |
+
"loss": 0.0075,
|
| 662 |
+
"step": 420
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 423.94,
|
| 666 |
+
"learning_rate": 0.001,
|
| 667 |
+
"loss": 0.0071,
|
| 668 |
+
"step": 424
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 427.94,
|
| 672 |
+
"learning_rate": 0.001,
|
| 673 |
+
"loss": 0.0072,
|
| 674 |
+
"step": 428
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 431.94,
|
| 678 |
+
"learning_rate": 0.001,
|
| 679 |
+
"loss": 0.0079,
|
| 680 |
+
"step": 432
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 435.94,
|
| 684 |
+
"learning_rate": 0.001,
|
| 685 |
+
"loss": 0.0288,
|
| 686 |
+
"step": 436
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 439.94,
|
| 690 |
+
"learning_rate": 0.001,
|
| 691 |
+
"loss": 0.0244,
|
| 692 |
+
"step": 440
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 443.94,
|
| 696 |
+
"learning_rate": 0.001,
|
| 697 |
+
"loss": 0.0254,
|
| 698 |
+
"step": 444
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 447.94,
|
| 702 |
+
"learning_rate": 0.001,
|
| 703 |
+
"loss": 0.0131,
|
| 704 |
+
"step": 448
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 451.94,
|
| 708 |
+
"learning_rate": 0.001,
|
| 709 |
+
"loss": 0.0096,
|
| 710 |
+
"step": 452
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 455.94,
|
| 714 |
+
"learning_rate": 0.001,
|
| 715 |
+
"loss": 0.0064,
|
| 716 |
+
"step": 456
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 459.94,
|
| 720 |
+
"learning_rate": 0.001,
|
| 721 |
+
"loss": 0.006,
|
| 722 |
+
"step": 460
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 463.94,
|
| 726 |
+
"learning_rate": 0.001,
|
| 727 |
+
"loss": 0.0058,
|
| 728 |
+
"step": 464
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 467.94,
|
| 732 |
+
"learning_rate": 0.001,
|
| 733 |
+
"loss": 0.0057,
|
| 734 |
+
"step": 468
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 471.94,
|
| 738 |
+
"learning_rate": 0.001,
|
| 739 |
+
"loss": 0.0056,
|
| 740 |
+
"step": 472
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 475.94,
|
| 744 |
+
"learning_rate": 0.001,
|
| 745 |
+
"loss": 0.0055,
|
| 746 |
+
"step": 476
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 479.94,
|
| 750 |
+
"learning_rate": 0.001,
|
| 751 |
+
"loss": 0.0053,
|
| 752 |
+
"step": 480
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 483.94,
|
| 756 |
+
"learning_rate": 0.001,
|
| 757 |
+
"loss": 0.0049,
|
| 758 |
+
"step": 484
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 487.94,
|
| 762 |
+
"learning_rate": 0.001,
|
| 763 |
+
"loss": 0.0048,
|
| 764 |
+
"step": 488
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 491.94,
|
| 768 |
+
"learning_rate": 0.001,
|
| 769 |
+
"loss": 0.0047,
|
| 770 |
+
"step": 492
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 495.94,
|
| 774 |
+
"learning_rate": 0.001,
|
| 775 |
+
"loss": 0.0048,
|
| 776 |
+
"step": 496
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 499.94,
|
| 780 |
+
"learning_rate": 0.001,
|
| 781 |
+
"loss": 0.0045,
|
| 782 |
+
"step": 500
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 503.94,
|
| 786 |
+
"learning_rate": 0.001,
|
| 787 |
+
"loss": 0.0045,
|
| 788 |
+
"step": 504
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 507.94,
|
| 792 |
+
"learning_rate": 0.001,
|
| 793 |
+
"loss": 0.0042,
|
| 794 |
+
"step": 508
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 511.94,
|
| 798 |
+
"learning_rate": 0.001,
|
| 799 |
+
"loss": 0.0044,
|
| 800 |
+
"step": 512
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 515.94,
|
| 804 |
+
"learning_rate": 0.001,
|
| 805 |
+
"loss": 0.0042,
|
| 806 |
+
"step": 516
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 519.94,
|
| 810 |
+
"learning_rate": 0.001,
|
| 811 |
+
"loss": 0.0039,
|
| 812 |
+
"step": 520
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 523.94,
|
| 816 |
+
"learning_rate": 0.001,
|
| 817 |
+
"loss": 0.0039,
|
| 818 |
+
"step": 524
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 527.94,
|
| 822 |
+
"learning_rate": 0.001,
|
| 823 |
+
"loss": 0.0039,
|
| 824 |
+
"step": 528
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 531.94,
|
| 828 |
+
"learning_rate": 0.001,
|
| 829 |
+
"loss": 0.0045,
|
| 830 |
+
"step": 532
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 535.94,
|
| 834 |
+
"learning_rate": 0.001,
|
| 835 |
+
"loss": 0.0223,
|
| 836 |
+
"step": 536
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 539.94,
|
| 840 |
+
"learning_rate": 0.001,
|
| 841 |
+
"loss": 0.0217,
|
| 842 |
+
"step": 540
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 543.94,
|
| 846 |
+
"learning_rate": 0.001,
|
| 847 |
+
"loss": 0.0182,
|
| 848 |
+
"step": 544
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 547.94,
|
| 852 |
+
"learning_rate": 0.001,
|
| 853 |
+
"loss": 0.0253,
|
| 854 |
+
"step": 548
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 551.94,
|
| 858 |
+
"learning_rate": 0.001,
|
| 859 |
+
"loss": 0.023,
|
| 860 |
+
"step": 552
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 555.94,
|
| 864 |
+
"learning_rate": 0.001,
|
| 865 |
+
"loss": 0.0738,
|
| 866 |
+
"step": 556
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 559.94,
|
| 870 |
+
"learning_rate": 0.001,
|
| 871 |
+
"loss": 0.0071,
|
| 872 |
+
"step": 560
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 563.94,
|
| 876 |
+
"learning_rate": 0.001,
|
| 877 |
+
"loss": 0.0049,
|
| 878 |
+
"step": 564
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 567.94,
|
| 882 |
+
"learning_rate": 0.001,
|
| 883 |
+
"loss": 0.0045,
|
| 884 |
+
"step": 568
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 571.94,
|
| 888 |
+
"learning_rate": 0.001,
|
| 889 |
+
"loss": 0.0041,
|
| 890 |
+
"step": 572
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 575.94,
|
| 894 |
+
"learning_rate": 0.001,
|
| 895 |
+
"loss": 0.004,
|
| 896 |
+
"step": 576
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 579.94,
|
| 900 |
+
"learning_rate": 0.001,
|
| 901 |
+
"loss": 0.0036,
|
| 902 |
+
"step": 580
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 583.94,
|
| 906 |
+
"learning_rate": 0.001,
|
| 907 |
+
"loss": 0.0035,
|
| 908 |
+
"step": 584
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 587.94,
|
| 912 |
+
"learning_rate": 0.001,
|
| 913 |
+
"loss": 0.0035,
|
| 914 |
+
"step": 588
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 591.94,
|
| 918 |
+
"learning_rate": 0.001,
|
| 919 |
+
"loss": 0.0032,
|
| 920 |
+
"step": 592
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 595.94,
|
| 924 |
+
"learning_rate": 0.001,
|
| 925 |
+
"loss": 0.0033,
|
| 926 |
+
"step": 596
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 599.94,
|
| 930 |
+
"learning_rate": 0.001,
|
| 931 |
+
"loss": 0.0032,
|
| 932 |
+
"step": 600
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 599.94,
|
| 936 |
+
"eval_exact_match": 0.47775628626692457,
|
| 937 |
+
"eval_exec": 0.4709864603481625,
|
| 938 |
+
"eval_loss": 0.5292267203330994,
|
| 939 |
+
"eval_runtime": 108.5346,
|
| 940 |
+
"eval_samples_per_second": 9.527,
|
| 941 |
+
"step": 600
|
| 942 |
+
}
|
| 943 |
+
],
|
| 944 |
+
"max_steps": 1602,
|
| 945 |
+
"num_train_epochs": 1602,
|
| 946 |
+
"total_flos": 6.865408617747579e+17,
|
| 947 |
+
"trial_name": null,
|
| 948 |
+
"trial_params": null
|
| 949 |
+
}
|
checkpoint-600/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c20e4fbfc3c6e458b39e5e8bb30e9452245ef3516a01db5a085e691f53119d55
|
| 3 |
+
size 3055
|