End of training
Browse files- README.md +26 -36
- added_tokens.json +102 -0
- config.json +1 -1
- generation_config.json +1 -1
- model.safetensors +3 -0
- runs/Feb09_09-32-14_53af0da22b49/events.out.tfevents.1707471138.53af0da22b49.7344.6 +3 -0
- special_tokens_map.json +21 -3
- tokenizer_config.json +826 -0
- trainer_state.json +924 -974
- training_args.bin +2 -2
README.md
CHANGED
|
@@ -15,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [alexue4/text-normalization-ru-new](https://huggingface.co/alexue4/text-normalization-ru-new) on the None dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 0.
|
| 19 |
- Mean Distance: 0
|
| 20 |
-
- Max Distance:
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
@@ -37,48 +37,38 @@ More information needed
|
|
| 37 |
|
| 38 |
The following hyperparameters were used during training:
|
| 39 |
- learning_rate: 0.0001
|
| 40 |
-
- train_batch_size:
|
| 41 |
-
- eval_batch_size:
|
| 42 |
- seed: 42
|
| 43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
- lr_scheduler_type: linear
|
| 45 |
- lr_scheduler_warmup_ratio: 0.1
|
| 46 |
-
- num_epochs:
|
| 47 |
|
| 48 |
### Training results
|
| 49 |
|
| 50 |
-
| Training Loss | Epoch | Step
|
| 51 |
-
|
| 52 |
-
| 0.
|
| 53 |
-
| 0.
|
| 54 |
-
| 0.
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
| 59 |
-
| 0.
|
| 60 |
-
| 0.
|
| 61 |
-
| 0.
|
| 62 |
-
| 0.
|
| 63 |
-
| 0.
|
| 64 |
-
| 0.
|
| 65 |
-
| 0.
|
| 66 |
-
| 0.
|
| 67 |
-
| 0.0009 | 16.0 | 1104 | 0.0007 | 0 | 3 |
|
| 68 |
-
| 0.0014 | 17.0 | 1173 | 0.0003 | 0 | 0 |
|
| 69 |
-
| 0.001 | 18.0 | 1242 | 0.0004 | 0 | 0 |
|
| 70 |
-
| 0.0007 | 19.0 | 1311 | 0.0013 | 0 | 3 |
|
| 71 |
-
| 0.0013 | 20.0 | 1380 | 0.0013 | 0 | 3 |
|
| 72 |
-
| 0.0007 | 21.0 | 1449 | 0.0003 | 0 | 0 |
|
| 73 |
-
| 0.0016 | 22.0 | 1518 | 0.0003 | 0 | 0 |
|
| 74 |
-
| 0.0013 | 23.0 | 1587 | 0.0003 | 0 | 0 |
|
| 75 |
-
| 0.0004 | 24.0 | 1656 | 0.0003 | 0 | 0 |
|
| 76 |
-
| 0.001 | 25.0 | 1725 | 0.0003 | 0 | 0 |
|
| 77 |
|
| 78 |
|
| 79 |
### Framework versions
|
| 80 |
|
| 81 |
-
- Transformers 4.
|
| 82 |
-
- Pytorch 2.0
|
| 83 |
-
- Datasets 2.
|
| 84 |
-
- Tokenizers 0.
|
|
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [alexue4/text-normalization-ru-new](https://huggingface.co/alexue4/text-normalization-ru-new) on the None dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.0008
|
| 19 |
- Mean Distance: 0
|
| 20 |
+
- Max Distance: 3
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 37 |
|
| 38 |
The following hyperparameters were used during training:
|
| 39 |
- learning_rate: 0.0001
|
| 40 |
+
- train_batch_size: 150
|
| 41 |
+
- eval_batch_size: 150
|
| 42 |
- seed: 42
|
| 43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
- lr_scheduler_type: linear
|
| 45 |
- lr_scheduler_warmup_ratio: 0.1
|
| 46 |
+
- num_epochs: 15
|
| 47 |
|
| 48 |
### Training results
|
| 49 |
|
| 50 |
+
| Training Loss | Epoch | Step | Validation Loss | Mean Distance | Max Distance |
|
| 51 |
+
|:-------------:|:-----:|:-----:|:---------------:|:-------------:|:------------:|
|
| 52 |
+
| 0.0004 | 1.0 | 3059 | 0.0022 | 0 | 12 |
|
| 53 |
+
| 0.0003 | 2.0 | 6118 | 0.0011 | 0 | 5 |
|
| 54 |
+
| 0.0003 | 3.0 | 9177 | 0.0010 | 0 | 6 |
|
| 55 |
+
| 0.0003 | 4.0 | 12236 | 0.0012 | 0 | 3 |
|
| 56 |
+
| 0.0003 | 5.0 | 15295 | 0.0008 | 0 | 3 |
|
| 57 |
+
| 0.0002 | 6.0 | 18354 | 0.0009 | 0 | 3 |
|
| 58 |
+
| 0.0002 | 7.0 | 21413 | 0.0008 | 0 | 3 |
|
| 59 |
+
| 0.0002 | 8.0 | 24472 | 0.0008 | 0 | 3 |
|
| 60 |
+
| 0.0002 | 9.0 | 27531 | 0.0007 | 0 | 3 |
|
| 61 |
+
| 0.0002 | 10.0 | 30590 | 0.0008 | 0 | 3 |
|
| 62 |
+
| 0.0002 | 11.0 | 33649 | 0.0008 | 0 | 3 |
|
| 63 |
+
| 0.0002 | 12.0 | 36708 | 0.0008 | 0 | 3 |
|
| 64 |
+
| 0.0002 | 13.0 | 39767 | 0.0008 | 0 | 3 |
|
| 65 |
+
| 0.0002 | 14.0 | 42826 | 0.0008 | 0 | 3 |
|
| 66 |
+
| 0.0002 | 15.0 | 45885 | 0.0008 | 0 | 3 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 67 |
|
| 68 |
|
| 69 |
### Framework versions
|
| 70 |
|
| 71 |
+
- Transformers 4.37.2
|
| 72 |
+
- Pytorch 2.1.0+cu121
|
| 73 |
+
- Datasets 2.16.1
|
| 74 |
+
- Tokenizers 0.15.1
|
added_tokens.json
ADDED
|
@@ -0,0 +1,102 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<extra_id_0>": 20199,
|
| 3 |
+
"<extra_id_10>": 20189,
|
| 4 |
+
"<extra_id_11>": 20188,
|
| 5 |
+
"<extra_id_12>": 20187,
|
| 6 |
+
"<extra_id_13>": 20186,
|
| 7 |
+
"<extra_id_14>": 20185,
|
| 8 |
+
"<extra_id_15>": 20184,
|
| 9 |
+
"<extra_id_16>": 20183,
|
| 10 |
+
"<extra_id_17>": 20182,
|
| 11 |
+
"<extra_id_18>": 20181,
|
| 12 |
+
"<extra_id_19>": 20180,
|
| 13 |
+
"<extra_id_1>": 20198,
|
| 14 |
+
"<extra_id_20>": 20179,
|
| 15 |
+
"<extra_id_21>": 20178,
|
| 16 |
+
"<extra_id_22>": 20177,
|
| 17 |
+
"<extra_id_23>": 20176,
|
| 18 |
+
"<extra_id_24>": 20175,
|
| 19 |
+
"<extra_id_25>": 20174,
|
| 20 |
+
"<extra_id_26>": 20173,
|
| 21 |
+
"<extra_id_27>": 20172,
|
| 22 |
+
"<extra_id_28>": 20171,
|
| 23 |
+
"<extra_id_29>": 20170,
|
| 24 |
+
"<extra_id_2>": 20197,
|
| 25 |
+
"<extra_id_30>": 20169,
|
| 26 |
+
"<extra_id_31>": 20168,
|
| 27 |
+
"<extra_id_32>": 20167,
|
| 28 |
+
"<extra_id_33>": 20166,
|
| 29 |
+
"<extra_id_34>": 20165,
|
| 30 |
+
"<extra_id_35>": 20164,
|
| 31 |
+
"<extra_id_36>": 20163,
|
| 32 |
+
"<extra_id_37>": 20162,
|
| 33 |
+
"<extra_id_38>": 20161,
|
| 34 |
+
"<extra_id_39>": 20160,
|
| 35 |
+
"<extra_id_3>": 20196,
|
| 36 |
+
"<extra_id_40>": 20159,
|
| 37 |
+
"<extra_id_41>": 20158,
|
| 38 |
+
"<extra_id_42>": 20157,
|
| 39 |
+
"<extra_id_43>": 20156,
|
| 40 |
+
"<extra_id_44>": 20155,
|
| 41 |
+
"<extra_id_45>": 20154,
|
| 42 |
+
"<extra_id_46>": 20153,
|
| 43 |
+
"<extra_id_47>": 20152,
|
| 44 |
+
"<extra_id_48>": 20151,
|
| 45 |
+
"<extra_id_49>": 20150,
|
| 46 |
+
"<extra_id_4>": 20195,
|
| 47 |
+
"<extra_id_50>": 20149,
|
| 48 |
+
"<extra_id_51>": 20148,
|
| 49 |
+
"<extra_id_52>": 20147,
|
| 50 |
+
"<extra_id_53>": 20146,
|
| 51 |
+
"<extra_id_54>": 20145,
|
| 52 |
+
"<extra_id_55>": 20144,
|
| 53 |
+
"<extra_id_56>": 20143,
|
| 54 |
+
"<extra_id_57>": 20142,
|
| 55 |
+
"<extra_id_58>": 20141,
|
| 56 |
+
"<extra_id_59>": 20140,
|
| 57 |
+
"<extra_id_5>": 20194,
|
| 58 |
+
"<extra_id_60>": 20139,
|
| 59 |
+
"<extra_id_61>": 20138,
|
| 60 |
+
"<extra_id_62>": 20137,
|
| 61 |
+
"<extra_id_63>": 20136,
|
| 62 |
+
"<extra_id_64>": 20135,
|
| 63 |
+
"<extra_id_65>": 20134,
|
| 64 |
+
"<extra_id_66>": 20133,
|
| 65 |
+
"<extra_id_67>": 20132,
|
| 66 |
+
"<extra_id_68>": 20131,
|
| 67 |
+
"<extra_id_69>": 20130,
|
| 68 |
+
"<extra_id_6>": 20193,
|
| 69 |
+
"<extra_id_70>": 20129,
|
| 70 |
+
"<extra_id_71>": 20128,
|
| 71 |
+
"<extra_id_72>": 20127,
|
| 72 |
+
"<extra_id_73>": 20126,
|
| 73 |
+
"<extra_id_74>": 20125,
|
| 74 |
+
"<extra_id_75>": 20124,
|
| 75 |
+
"<extra_id_76>": 20123,
|
| 76 |
+
"<extra_id_77>": 20122,
|
| 77 |
+
"<extra_id_78>": 20121,
|
| 78 |
+
"<extra_id_79>": 20120,
|
| 79 |
+
"<extra_id_7>": 20192,
|
| 80 |
+
"<extra_id_80>": 20119,
|
| 81 |
+
"<extra_id_81>": 20118,
|
| 82 |
+
"<extra_id_82>": 20117,
|
| 83 |
+
"<extra_id_83>": 20116,
|
| 84 |
+
"<extra_id_84>": 20115,
|
| 85 |
+
"<extra_id_85>": 20114,
|
| 86 |
+
"<extra_id_86>": 20113,
|
| 87 |
+
"<extra_id_87>": 20112,
|
| 88 |
+
"<extra_id_88>": 20111,
|
| 89 |
+
"<extra_id_89>": 20110,
|
| 90 |
+
"<extra_id_8>": 20191,
|
| 91 |
+
"<extra_id_90>": 20109,
|
| 92 |
+
"<extra_id_91>": 20108,
|
| 93 |
+
"<extra_id_92>": 20107,
|
| 94 |
+
"<extra_id_93>": 20106,
|
| 95 |
+
"<extra_id_94>": 20105,
|
| 96 |
+
"<extra_id_95>": 20104,
|
| 97 |
+
"<extra_id_96>": 20103,
|
| 98 |
+
"<extra_id_97>": 20102,
|
| 99 |
+
"<extra_id_98>": 20101,
|
| 100 |
+
"<extra_id_99>": 20100,
|
| 101 |
+
"<extra_id_9>": 20190
|
| 102 |
+
}
|
config.json
CHANGED
|
@@ -26,7 +26,7 @@
|
|
| 26 |
"tie_word_embeddings": false,
|
| 27 |
"tokenizer_class": "T5Tokenizer",
|
| 28 |
"torch_dtype": "float32",
|
| 29 |
-
"transformers_version": "4.
|
| 30 |
"use_cache": true,
|
| 31 |
"vocab_size": 20100
|
| 32 |
}
|
|
|
|
| 26 |
"tie_word_embeddings": false,
|
| 27 |
"tokenizer_class": "T5Tokenizer",
|
| 28 |
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.37.2",
|
| 30 |
"use_cache": true,
|
| 31 |
"vocab_size": 20100
|
| 32 |
}
|
generation_config.json
CHANGED
|
@@ -3,5 +3,5 @@
|
|
| 3 |
"decoder_start_token_id": 0,
|
| 4 |
"eos_token_id": 1,
|
| 5 |
"pad_token_id": 0,
|
| 6 |
-
"transformers_version": "4.
|
| 7 |
}
|
|
|
|
| 3 |
"decoder_start_token_id": 0,
|
| 4 |
"eos_token_id": 1,
|
| 5 |
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.37.2"
|
| 7 |
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6110aaf653c8f40296e4b2ecfada2f35edce407a76dc3ea7e879357e5b60b1fb
|
| 3 |
+
size 258600360
|
runs/Feb09_09-32-14_53af0da22b49/events.out.tfevents.1707471138.53af0da22b49.7344.6
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2710126dbff402c0f7c727617f35948c443e13334be9e4d553f6fd3f73aa4b19
|
| 3 |
+
size 47147
|
special_tokens_map.json
CHANGED
|
@@ -101,7 +101,25 @@
|
|
| 101 |
"<extra_id_98>",
|
| 102 |
"<extra_id_99>"
|
| 103 |
],
|
| 104 |
-
"eos_token":
|
| 105 |
-
|
| 106 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 107 |
}
|
|
|
|
| 101 |
"<extra_id_98>",
|
| 102 |
"<extra_id_99>"
|
| 103 |
],
|
| 104 |
+
"eos_token": {
|
| 105 |
+
"content": "</s>",
|
| 106 |
+
"lstrip": false,
|
| 107 |
+
"normalized": false,
|
| 108 |
+
"rstrip": false,
|
| 109 |
+
"single_word": false
|
| 110 |
+
},
|
| 111 |
+
"pad_token": {
|
| 112 |
+
"content": "<pad>",
|
| 113 |
+
"lstrip": false,
|
| 114 |
+
"normalized": false,
|
| 115 |
+
"rstrip": false,
|
| 116 |
+
"single_word": false
|
| 117 |
+
},
|
| 118 |
+
"unk_token": {
|
| 119 |
+
"content": "<unk>",
|
| 120 |
+
"lstrip": false,
|
| 121 |
+
"normalized": false,
|
| 122 |
+
"rstrip": false,
|
| 123 |
+
"single_word": false
|
| 124 |
+
}
|
| 125 |
}
|
tokenizer_config.json
CHANGED
|
@@ -1,4 +1,830 @@
|
|
| 1 |
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
"additional_special_tokens": [
|
| 3 |
"<extra_id_0>",
|
| 4 |
"<extra_id_1>",
|
|
|
|
| 1 |
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<pad>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "</s>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "<unk>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"20100": {
|
| 28 |
+
"content": "<extra_id_99>",
|
| 29 |
+
"lstrip": true,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": true,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"20101": {
|
| 36 |
+
"content": "<extra_id_98>",
|
| 37 |
+
"lstrip": true,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": true,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"20102": {
|
| 44 |
+
"content": "<extra_id_97>",
|
| 45 |
+
"lstrip": true,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": true,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"20103": {
|
| 52 |
+
"content": "<extra_id_96>",
|
| 53 |
+
"lstrip": true,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": true,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"20104": {
|
| 60 |
+
"content": "<extra_id_95>",
|
| 61 |
+
"lstrip": true,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": true,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"20105": {
|
| 68 |
+
"content": "<extra_id_94>",
|
| 69 |
+
"lstrip": true,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": true,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"20106": {
|
| 76 |
+
"content": "<extra_id_93>",
|
| 77 |
+
"lstrip": true,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": true,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
"20107": {
|
| 84 |
+
"content": "<extra_id_92>",
|
| 85 |
+
"lstrip": true,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": true,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": true
|
| 90 |
+
},
|
| 91 |
+
"20108": {
|
| 92 |
+
"content": "<extra_id_91>",
|
| 93 |
+
"lstrip": true,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": true,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": true
|
| 98 |
+
},
|
| 99 |
+
"20109": {
|
| 100 |
+
"content": "<extra_id_90>",
|
| 101 |
+
"lstrip": true,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": true,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": true
|
| 106 |
+
},
|
| 107 |
+
"20110": {
|
| 108 |
+
"content": "<extra_id_89>",
|
| 109 |
+
"lstrip": true,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": true,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": true
|
| 114 |
+
},
|
| 115 |
+
"20111": {
|
| 116 |
+
"content": "<extra_id_88>",
|
| 117 |
+
"lstrip": true,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": true,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"20112": {
|
| 124 |
+
"content": "<extra_id_87>",
|
| 125 |
+
"lstrip": true,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": true,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"20113": {
|
| 132 |
+
"content": "<extra_id_86>",
|
| 133 |
+
"lstrip": true,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": true,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"20114": {
|
| 140 |
+
"content": "<extra_id_85>",
|
| 141 |
+
"lstrip": true,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": true,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"20115": {
|
| 148 |
+
"content": "<extra_id_84>",
|
| 149 |
+
"lstrip": true,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": true,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"20116": {
|
| 156 |
+
"content": "<extra_id_83>",
|
| 157 |
+
"lstrip": true,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": true,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": true
|
| 162 |
+
},
|
| 163 |
+
"20117": {
|
| 164 |
+
"content": "<extra_id_82>",
|
| 165 |
+
"lstrip": true,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": true,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": true
|
| 170 |
+
},
|
| 171 |
+
"20118": {
|
| 172 |
+
"content": "<extra_id_81>",
|
| 173 |
+
"lstrip": true,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": true,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
},
|
| 179 |
+
"20119": {
|
| 180 |
+
"content": "<extra_id_80>",
|
| 181 |
+
"lstrip": true,
|
| 182 |
+
"normalized": false,
|
| 183 |
+
"rstrip": true,
|
| 184 |
+
"single_word": false,
|
| 185 |
+
"special": true
|
| 186 |
+
},
|
| 187 |
+
"20120": {
|
| 188 |
+
"content": "<extra_id_79>",
|
| 189 |
+
"lstrip": true,
|
| 190 |
+
"normalized": false,
|
| 191 |
+
"rstrip": true,
|
| 192 |
+
"single_word": false,
|
| 193 |
+
"special": true
|
| 194 |
+
},
|
| 195 |
+
"20121": {
|
| 196 |
+
"content": "<extra_id_78>",
|
| 197 |
+
"lstrip": true,
|
| 198 |
+
"normalized": false,
|
| 199 |
+
"rstrip": true,
|
| 200 |
+
"single_word": false,
|
| 201 |
+
"special": true
|
| 202 |
+
},
|
| 203 |
+
"20122": {
|
| 204 |
+
"content": "<extra_id_77>",
|
| 205 |
+
"lstrip": true,
|
| 206 |
+
"normalized": false,
|
| 207 |
+
"rstrip": true,
|
| 208 |
+
"single_word": false,
|
| 209 |
+
"special": true
|
| 210 |
+
},
|
| 211 |
+
"20123": {
|
| 212 |
+
"content": "<extra_id_76>",
|
| 213 |
+
"lstrip": true,
|
| 214 |
+
"normalized": false,
|
| 215 |
+
"rstrip": true,
|
| 216 |
+
"single_word": false,
|
| 217 |
+
"special": true
|
| 218 |
+
},
|
| 219 |
+
"20124": {
|
| 220 |
+
"content": "<extra_id_75>",
|
| 221 |
+
"lstrip": true,
|
| 222 |
+
"normalized": false,
|
| 223 |
+
"rstrip": true,
|
| 224 |
+
"single_word": false,
|
| 225 |
+
"special": true
|
| 226 |
+
},
|
| 227 |
+
"20125": {
|
| 228 |
+
"content": "<extra_id_74>",
|
| 229 |
+
"lstrip": true,
|
| 230 |
+
"normalized": false,
|
| 231 |
+
"rstrip": true,
|
| 232 |
+
"single_word": false,
|
| 233 |
+
"special": true
|
| 234 |
+
},
|
| 235 |
+
"20126": {
|
| 236 |
+
"content": "<extra_id_73>",
|
| 237 |
+
"lstrip": true,
|
| 238 |
+
"normalized": false,
|
| 239 |
+
"rstrip": true,
|
| 240 |
+
"single_word": false,
|
| 241 |
+
"special": true
|
| 242 |
+
},
|
| 243 |
+
"20127": {
|
| 244 |
+
"content": "<extra_id_72>",
|
| 245 |
+
"lstrip": true,
|
| 246 |
+
"normalized": false,
|
| 247 |
+
"rstrip": true,
|
| 248 |
+
"single_word": false,
|
| 249 |
+
"special": true
|
| 250 |
+
},
|
| 251 |
+
"20128": {
|
| 252 |
+
"content": "<extra_id_71>",
|
| 253 |
+
"lstrip": true,
|
| 254 |
+
"normalized": false,
|
| 255 |
+
"rstrip": true,
|
| 256 |
+
"single_word": false,
|
| 257 |
+
"special": true
|
| 258 |
+
},
|
| 259 |
+
"20129": {
|
| 260 |
+
"content": "<extra_id_70>",
|
| 261 |
+
"lstrip": true,
|
| 262 |
+
"normalized": false,
|
| 263 |
+
"rstrip": true,
|
| 264 |
+
"single_word": false,
|
| 265 |
+
"special": true
|
| 266 |
+
},
|
| 267 |
+
"20130": {
|
| 268 |
+
"content": "<extra_id_69>",
|
| 269 |
+
"lstrip": true,
|
| 270 |
+
"normalized": false,
|
| 271 |
+
"rstrip": true,
|
| 272 |
+
"single_word": false,
|
| 273 |
+
"special": true
|
| 274 |
+
},
|
| 275 |
+
"20131": {
|
| 276 |
+
"content": "<extra_id_68>",
|
| 277 |
+
"lstrip": true,
|
| 278 |
+
"normalized": false,
|
| 279 |
+
"rstrip": true,
|
| 280 |
+
"single_word": false,
|
| 281 |
+
"special": true
|
| 282 |
+
},
|
| 283 |
+
"20132": {
|
| 284 |
+
"content": "<extra_id_67>",
|
| 285 |
+
"lstrip": true,
|
| 286 |
+
"normalized": false,
|
| 287 |
+
"rstrip": true,
|
| 288 |
+
"single_word": false,
|
| 289 |
+
"special": true
|
| 290 |
+
},
|
| 291 |
+
"20133": {
|
| 292 |
+
"content": "<extra_id_66>",
|
| 293 |
+
"lstrip": true,
|
| 294 |
+
"normalized": false,
|
| 295 |
+
"rstrip": true,
|
| 296 |
+
"single_word": false,
|
| 297 |
+
"special": true
|
| 298 |
+
},
|
| 299 |
+
"20134": {
|
| 300 |
+
"content": "<extra_id_65>",
|
| 301 |
+
"lstrip": true,
|
| 302 |
+
"normalized": false,
|
| 303 |
+
"rstrip": true,
|
| 304 |
+
"single_word": false,
|
| 305 |
+
"special": true
|
| 306 |
+
},
|
| 307 |
+
"20135": {
|
| 308 |
+
"content": "<extra_id_64>",
|
| 309 |
+
"lstrip": true,
|
| 310 |
+
"normalized": false,
|
| 311 |
+
"rstrip": true,
|
| 312 |
+
"single_word": false,
|
| 313 |
+
"special": true
|
| 314 |
+
},
|
| 315 |
+
"20136": {
|
| 316 |
+
"content": "<extra_id_63>",
|
| 317 |
+
"lstrip": true,
|
| 318 |
+
"normalized": false,
|
| 319 |
+
"rstrip": true,
|
| 320 |
+
"single_word": false,
|
| 321 |
+
"special": true
|
| 322 |
+
},
|
| 323 |
+
"20137": {
|
| 324 |
+
"content": "<extra_id_62>",
|
| 325 |
+
"lstrip": true,
|
| 326 |
+
"normalized": false,
|
| 327 |
+
"rstrip": true,
|
| 328 |
+
"single_word": false,
|
| 329 |
+
"special": true
|
| 330 |
+
},
|
| 331 |
+
"20138": {
|
| 332 |
+
"content": "<extra_id_61>",
|
| 333 |
+
"lstrip": true,
|
| 334 |
+
"normalized": false,
|
| 335 |
+
"rstrip": true,
|
| 336 |
+
"single_word": false,
|
| 337 |
+
"special": true
|
| 338 |
+
},
|
| 339 |
+
"20139": {
|
| 340 |
+
"content": "<extra_id_60>",
|
| 341 |
+
"lstrip": true,
|
| 342 |
+
"normalized": false,
|
| 343 |
+
"rstrip": true,
|
| 344 |
+
"single_word": false,
|
| 345 |
+
"special": true
|
| 346 |
+
},
|
| 347 |
+
"20140": {
|
| 348 |
+
"content": "<extra_id_59>",
|
| 349 |
+
"lstrip": true,
|
| 350 |
+
"normalized": false,
|
| 351 |
+
"rstrip": true,
|
| 352 |
+
"single_word": false,
|
| 353 |
+
"special": true
|
| 354 |
+
},
|
| 355 |
+
"20141": {
|
| 356 |
+
"content": "<extra_id_58>",
|
| 357 |
+
"lstrip": true,
|
| 358 |
+
"normalized": false,
|
| 359 |
+
"rstrip": true,
|
| 360 |
+
"single_word": false,
|
| 361 |
+
"special": true
|
| 362 |
+
},
|
| 363 |
+
"20142": {
|
| 364 |
+
"content": "<extra_id_57>",
|
| 365 |
+
"lstrip": true,
|
| 366 |
+
"normalized": false,
|
| 367 |
+
"rstrip": true,
|
| 368 |
+
"single_word": false,
|
| 369 |
+
"special": true
|
| 370 |
+
},
|
| 371 |
+
"20143": {
|
| 372 |
+
"content": "<extra_id_56>",
|
| 373 |
+
"lstrip": true,
|
| 374 |
+
"normalized": false,
|
| 375 |
+
"rstrip": true,
|
| 376 |
+
"single_word": false,
|
| 377 |
+
"special": true
|
| 378 |
+
},
|
| 379 |
+
"20144": {
|
| 380 |
+
"content": "<extra_id_55>",
|
| 381 |
+
"lstrip": true,
|
| 382 |
+
"normalized": false,
|
| 383 |
+
"rstrip": true,
|
| 384 |
+
"single_word": false,
|
| 385 |
+
"special": true
|
| 386 |
+
},
|
| 387 |
+
"20145": {
|
| 388 |
+
"content": "<extra_id_54>",
|
| 389 |
+
"lstrip": true,
|
| 390 |
+
"normalized": false,
|
| 391 |
+
"rstrip": true,
|
| 392 |
+
"single_word": false,
|
| 393 |
+
"special": true
|
| 394 |
+
},
|
| 395 |
+
"20146": {
|
| 396 |
+
"content": "<extra_id_53>",
|
| 397 |
+
"lstrip": true,
|
| 398 |
+
"normalized": false,
|
| 399 |
+
"rstrip": true,
|
| 400 |
+
"single_word": false,
|
| 401 |
+
"special": true
|
| 402 |
+
},
|
| 403 |
+
"20147": {
|
| 404 |
+
"content": "<extra_id_52>",
|
| 405 |
+
"lstrip": true,
|
| 406 |
+
"normalized": false,
|
| 407 |
+
"rstrip": true,
|
| 408 |
+
"single_word": false,
|
| 409 |
+
"special": true
|
| 410 |
+
},
|
| 411 |
+
"20148": {
|
| 412 |
+
"content": "<extra_id_51>",
|
| 413 |
+
"lstrip": true,
|
| 414 |
+
"normalized": false,
|
| 415 |
+
"rstrip": true,
|
| 416 |
+
"single_word": false,
|
| 417 |
+
"special": true
|
| 418 |
+
},
|
| 419 |
+
"20149": {
|
| 420 |
+
"content": "<extra_id_50>",
|
| 421 |
+
"lstrip": true,
|
| 422 |
+
"normalized": false,
|
| 423 |
+
"rstrip": true,
|
| 424 |
+
"single_word": false,
|
| 425 |
+
"special": true
|
| 426 |
+
},
|
| 427 |
+
"20150": {
|
| 428 |
+
"content": "<extra_id_49>",
|
| 429 |
+
"lstrip": true,
|
| 430 |
+
"normalized": false,
|
| 431 |
+
"rstrip": true,
|
| 432 |
+
"single_word": false,
|
| 433 |
+
"special": true
|
| 434 |
+
},
|
| 435 |
+
"20151": {
|
| 436 |
+
"content": "<extra_id_48>",
|
| 437 |
+
"lstrip": true,
|
| 438 |
+
"normalized": false,
|
| 439 |
+
"rstrip": true,
|
| 440 |
+
"single_word": false,
|
| 441 |
+
"special": true
|
| 442 |
+
},
|
| 443 |
+
"20152": {
|
| 444 |
+
"content": "<extra_id_47>",
|
| 445 |
+
"lstrip": true,
|
| 446 |
+
"normalized": false,
|
| 447 |
+
"rstrip": true,
|
| 448 |
+
"single_word": false,
|
| 449 |
+
"special": true
|
| 450 |
+
},
|
| 451 |
+
"20153": {
|
| 452 |
+
"content": "<extra_id_46>",
|
| 453 |
+
"lstrip": true,
|
| 454 |
+
"normalized": false,
|
| 455 |
+
"rstrip": true,
|
| 456 |
+
"single_word": false,
|
| 457 |
+
"special": true
|
| 458 |
+
},
|
| 459 |
+
"20154": {
|
| 460 |
+
"content": "<extra_id_45>",
|
| 461 |
+
"lstrip": true,
|
| 462 |
+
"normalized": false,
|
| 463 |
+
"rstrip": true,
|
| 464 |
+
"single_word": false,
|
| 465 |
+
"special": true
|
| 466 |
+
},
|
| 467 |
+
"20155": {
|
| 468 |
+
"content": "<extra_id_44>",
|
| 469 |
+
"lstrip": true,
|
| 470 |
+
"normalized": false,
|
| 471 |
+
"rstrip": true,
|
| 472 |
+
"single_word": false,
|
| 473 |
+
"special": true
|
| 474 |
+
},
|
| 475 |
+
"20156": {
|
| 476 |
+
"content": "<extra_id_43>",
|
| 477 |
+
"lstrip": true,
|
| 478 |
+
"normalized": false,
|
| 479 |
+
"rstrip": true,
|
| 480 |
+
"single_word": false,
|
| 481 |
+
"special": true
|
| 482 |
+
},
|
| 483 |
+
"20157": {
|
| 484 |
+
"content": "<extra_id_42>",
|
| 485 |
+
"lstrip": true,
|
| 486 |
+
"normalized": false,
|
| 487 |
+
"rstrip": true,
|
| 488 |
+
"single_word": false,
|
| 489 |
+
"special": true
|
| 490 |
+
},
|
| 491 |
+
"20158": {
|
| 492 |
+
"content": "<extra_id_41>",
|
| 493 |
+
"lstrip": true,
|
| 494 |
+
"normalized": false,
|
| 495 |
+
"rstrip": true,
|
| 496 |
+
"single_word": false,
|
| 497 |
+
"special": true
|
| 498 |
+
},
|
| 499 |
+
"20159": {
|
| 500 |
+
"content": "<extra_id_40>",
|
| 501 |
+
"lstrip": true,
|
| 502 |
+
"normalized": false,
|
| 503 |
+
"rstrip": true,
|
| 504 |
+
"single_word": false,
|
| 505 |
+
"special": true
|
| 506 |
+
},
|
| 507 |
+
"20160": {
|
| 508 |
+
"content": "<extra_id_39>",
|
| 509 |
+
"lstrip": true,
|
| 510 |
+
"normalized": false,
|
| 511 |
+
"rstrip": true,
|
| 512 |
+
"single_word": false,
|
| 513 |
+
"special": true
|
| 514 |
+
},
|
| 515 |
+
"20161": {
|
| 516 |
+
"content": "<extra_id_38>",
|
| 517 |
+
"lstrip": true,
|
| 518 |
+
"normalized": false,
|
| 519 |
+
"rstrip": true,
|
| 520 |
+
"single_word": false,
|
| 521 |
+
"special": true
|
| 522 |
+
},
|
| 523 |
+
"20162": {
|
| 524 |
+
"content": "<extra_id_37>",
|
| 525 |
+
"lstrip": true,
|
| 526 |
+
"normalized": false,
|
| 527 |
+
"rstrip": true,
|
| 528 |
+
"single_word": false,
|
| 529 |
+
"special": true
|
| 530 |
+
},
|
| 531 |
+
"20163": {
|
| 532 |
+
"content": "<extra_id_36>",
|
| 533 |
+
"lstrip": true,
|
| 534 |
+
"normalized": false,
|
| 535 |
+
"rstrip": true,
|
| 536 |
+
"single_word": false,
|
| 537 |
+
"special": true
|
| 538 |
+
},
|
| 539 |
+
"20164": {
|
| 540 |
+
"content": "<extra_id_35>",
|
| 541 |
+
"lstrip": true,
|
| 542 |
+
"normalized": false,
|
| 543 |
+
"rstrip": true,
|
| 544 |
+
"single_word": false,
|
| 545 |
+
"special": true
|
| 546 |
+
},
|
| 547 |
+
"20165": {
|
| 548 |
+
"content": "<extra_id_34>",
|
| 549 |
+
"lstrip": true,
|
| 550 |
+
"normalized": false,
|
| 551 |
+
"rstrip": true,
|
| 552 |
+
"single_word": false,
|
| 553 |
+
"special": true
|
| 554 |
+
},
|
| 555 |
+
"20166": {
|
| 556 |
+
"content": "<extra_id_33>",
|
| 557 |
+
"lstrip": true,
|
| 558 |
+
"normalized": false,
|
| 559 |
+
"rstrip": true,
|
| 560 |
+
"single_word": false,
|
| 561 |
+
"special": true
|
| 562 |
+
},
|
| 563 |
+
"20167": {
|
| 564 |
+
"content": "<extra_id_32>",
|
| 565 |
+
"lstrip": true,
|
| 566 |
+
"normalized": false,
|
| 567 |
+
"rstrip": true,
|
| 568 |
+
"single_word": false,
|
| 569 |
+
"special": true
|
| 570 |
+
},
|
| 571 |
+
"20168": {
|
| 572 |
+
"content": "<extra_id_31>",
|
| 573 |
+
"lstrip": true,
|
| 574 |
+
"normalized": false,
|
| 575 |
+
"rstrip": true,
|
| 576 |
+
"single_word": false,
|
| 577 |
+
"special": true
|
| 578 |
+
},
|
| 579 |
+
"20169": {
|
| 580 |
+
"content": "<extra_id_30>",
|
| 581 |
+
"lstrip": true,
|
| 582 |
+
"normalized": false,
|
| 583 |
+
"rstrip": true,
|
| 584 |
+
"single_word": false,
|
| 585 |
+
"special": true
|
| 586 |
+
},
|
| 587 |
+
"20170": {
|
| 588 |
+
"content": "<extra_id_29>",
|
| 589 |
+
"lstrip": true,
|
| 590 |
+
"normalized": false,
|
| 591 |
+
"rstrip": true,
|
| 592 |
+
"single_word": false,
|
| 593 |
+
"special": true
|
| 594 |
+
},
|
| 595 |
+
"20171": {
|
| 596 |
+
"content": "<extra_id_28>",
|
| 597 |
+
"lstrip": true,
|
| 598 |
+
"normalized": false,
|
| 599 |
+
"rstrip": true,
|
| 600 |
+
"single_word": false,
|
| 601 |
+
"special": true
|
| 602 |
+
},
|
| 603 |
+
"20172": {
|
| 604 |
+
"content": "<extra_id_27>",
|
| 605 |
+
"lstrip": true,
|
| 606 |
+
"normalized": false,
|
| 607 |
+
"rstrip": true,
|
| 608 |
+
"single_word": false,
|
| 609 |
+
"special": true
|
| 610 |
+
},
|
| 611 |
+
"20173": {
|
| 612 |
+
"content": "<extra_id_26>",
|
| 613 |
+
"lstrip": true,
|
| 614 |
+
"normalized": false,
|
| 615 |
+
"rstrip": true,
|
| 616 |
+
"single_word": false,
|
| 617 |
+
"special": true
|
| 618 |
+
},
|
| 619 |
+
"20174": {
|
| 620 |
+
"content": "<extra_id_25>",
|
| 621 |
+
"lstrip": true,
|
| 622 |
+
"normalized": false,
|
| 623 |
+
"rstrip": true,
|
| 624 |
+
"single_word": false,
|
| 625 |
+
"special": true
|
| 626 |
+
},
|
| 627 |
+
"20175": {
|
| 628 |
+
"content": "<extra_id_24>",
|
| 629 |
+
"lstrip": true,
|
| 630 |
+
"normalized": false,
|
| 631 |
+
"rstrip": true,
|
| 632 |
+
"single_word": false,
|
| 633 |
+
"special": true
|
| 634 |
+
},
|
| 635 |
+
"20176": {
|
| 636 |
+
"content": "<extra_id_23>",
|
| 637 |
+
"lstrip": true,
|
| 638 |
+
"normalized": false,
|
| 639 |
+
"rstrip": true,
|
| 640 |
+
"single_word": false,
|
| 641 |
+
"special": true
|
| 642 |
+
},
|
| 643 |
+
"20177": {
|
| 644 |
+
"content": "<extra_id_22>",
|
| 645 |
+
"lstrip": true,
|
| 646 |
+
"normalized": false,
|
| 647 |
+
"rstrip": true,
|
| 648 |
+
"single_word": false,
|
| 649 |
+
"special": true
|
| 650 |
+
},
|
| 651 |
+
"20178": {
|
| 652 |
+
"content": "<extra_id_21>",
|
| 653 |
+
"lstrip": true,
|
| 654 |
+
"normalized": false,
|
| 655 |
+
"rstrip": true,
|
| 656 |
+
"single_word": false,
|
| 657 |
+
"special": true
|
| 658 |
+
},
|
| 659 |
+
"20179": {
|
| 660 |
+
"content": "<extra_id_20>",
|
| 661 |
+
"lstrip": true,
|
| 662 |
+
"normalized": false,
|
| 663 |
+
"rstrip": true,
|
| 664 |
+
"single_word": false,
|
| 665 |
+
"special": true
|
| 666 |
+
},
|
| 667 |
+
"20180": {
|
| 668 |
+
"content": "<extra_id_19>",
|
| 669 |
+
"lstrip": true,
|
| 670 |
+
"normalized": false,
|
| 671 |
+
"rstrip": true,
|
| 672 |
+
"single_word": false,
|
| 673 |
+
"special": true
|
| 674 |
+
},
|
| 675 |
+
"20181": {
|
| 676 |
+
"content": "<extra_id_18>",
|
| 677 |
+
"lstrip": true,
|
| 678 |
+
"normalized": false,
|
| 679 |
+
"rstrip": true,
|
| 680 |
+
"single_word": false,
|
| 681 |
+
"special": true
|
| 682 |
+
},
|
| 683 |
+
"20182": {
|
| 684 |
+
"content": "<extra_id_17>",
|
| 685 |
+
"lstrip": true,
|
| 686 |
+
"normalized": false,
|
| 687 |
+
"rstrip": true,
|
| 688 |
+
"single_word": false,
|
| 689 |
+
"special": true
|
| 690 |
+
},
|
| 691 |
+
"20183": {
|
| 692 |
+
"content": "<extra_id_16>",
|
| 693 |
+
"lstrip": true,
|
| 694 |
+
"normalized": false,
|
| 695 |
+
"rstrip": true,
|
| 696 |
+
"single_word": false,
|
| 697 |
+
"special": true
|
| 698 |
+
},
|
| 699 |
+
"20184": {
|
| 700 |
+
"content": "<extra_id_15>",
|
| 701 |
+
"lstrip": true,
|
| 702 |
+
"normalized": false,
|
| 703 |
+
"rstrip": true,
|
| 704 |
+
"single_word": false,
|
| 705 |
+
"special": true
|
| 706 |
+
},
|
| 707 |
+
"20185": {
|
| 708 |
+
"content": "<extra_id_14>",
|
| 709 |
+
"lstrip": true,
|
| 710 |
+
"normalized": false,
|
| 711 |
+
"rstrip": true,
|
| 712 |
+
"single_word": false,
|
| 713 |
+
"special": true
|
| 714 |
+
},
|
| 715 |
+
"20186": {
|
| 716 |
+
"content": "<extra_id_13>",
|
| 717 |
+
"lstrip": true,
|
| 718 |
+
"normalized": false,
|
| 719 |
+
"rstrip": true,
|
| 720 |
+
"single_word": false,
|
| 721 |
+
"special": true
|
| 722 |
+
},
|
| 723 |
+
"20187": {
|
| 724 |
+
"content": "<extra_id_12>",
|
| 725 |
+
"lstrip": true,
|
| 726 |
+
"normalized": false,
|
| 727 |
+
"rstrip": true,
|
| 728 |
+
"single_word": false,
|
| 729 |
+
"special": true
|
| 730 |
+
},
|
| 731 |
+
"20188": {
|
| 732 |
+
"content": "<extra_id_11>",
|
| 733 |
+
"lstrip": true,
|
| 734 |
+
"normalized": false,
|
| 735 |
+
"rstrip": true,
|
| 736 |
+
"single_word": false,
|
| 737 |
+
"special": true
|
| 738 |
+
},
|
| 739 |
+
"20189": {
|
| 740 |
+
"content": "<extra_id_10>",
|
| 741 |
+
"lstrip": true,
|
| 742 |
+
"normalized": false,
|
| 743 |
+
"rstrip": true,
|
| 744 |
+
"single_word": false,
|
| 745 |
+
"special": true
|
| 746 |
+
},
|
| 747 |
+
"20190": {
|
| 748 |
+
"content": "<extra_id_9>",
|
| 749 |
+
"lstrip": true,
|
| 750 |
+
"normalized": false,
|
| 751 |
+
"rstrip": true,
|
| 752 |
+
"single_word": false,
|
| 753 |
+
"special": true
|
| 754 |
+
},
|
| 755 |
+
"20191": {
|
| 756 |
+
"content": "<extra_id_8>",
|
| 757 |
+
"lstrip": true,
|
| 758 |
+
"normalized": false,
|
| 759 |
+
"rstrip": true,
|
| 760 |
+
"single_word": false,
|
| 761 |
+
"special": true
|
| 762 |
+
},
|
| 763 |
+
"20192": {
|
| 764 |
+
"content": "<extra_id_7>",
|
| 765 |
+
"lstrip": true,
|
| 766 |
+
"normalized": false,
|
| 767 |
+
"rstrip": true,
|
| 768 |
+
"single_word": false,
|
| 769 |
+
"special": true
|
| 770 |
+
},
|
| 771 |
+
"20193": {
|
| 772 |
+
"content": "<extra_id_6>",
|
| 773 |
+
"lstrip": true,
|
| 774 |
+
"normalized": false,
|
| 775 |
+
"rstrip": true,
|
| 776 |
+
"single_word": false,
|
| 777 |
+
"special": true
|
| 778 |
+
},
|
| 779 |
+
"20194": {
|
| 780 |
+
"content": "<extra_id_5>",
|
| 781 |
+
"lstrip": true,
|
| 782 |
+
"normalized": false,
|
| 783 |
+
"rstrip": true,
|
| 784 |
+
"single_word": false,
|
| 785 |
+
"special": true
|
| 786 |
+
},
|
| 787 |
+
"20195": {
|
| 788 |
+
"content": "<extra_id_4>",
|
| 789 |
+
"lstrip": true,
|
| 790 |
+
"normalized": false,
|
| 791 |
+
"rstrip": true,
|
| 792 |
+
"single_word": false,
|
| 793 |
+
"special": true
|
| 794 |
+
},
|
| 795 |
+
"20196": {
|
| 796 |
+
"content": "<extra_id_3>",
|
| 797 |
+
"lstrip": true,
|
| 798 |
+
"normalized": false,
|
| 799 |
+
"rstrip": true,
|
| 800 |
+
"single_word": false,
|
| 801 |
+
"special": true
|
| 802 |
+
},
|
| 803 |
+
"20197": {
|
| 804 |
+
"content": "<extra_id_2>",
|
| 805 |
+
"lstrip": true,
|
| 806 |
+
"normalized": false,
|
| 807 |
+
"rstrip": true,
|
| 808 |
+
"single_word": false,
|
| 809 |
+
"special": true
|
| 810 |
+
},
|
| 811 |
+
"20198": {
|
| 812 |
+
"content": "<extra_id_1>",
|
| 813 |
+
"lstrip": true,
|
| 814 |
+
"normalized": false,
|
| 815 |
+
"rstrip": true,
|
| 816 |
+
"single_word": false,
|
| 817 |
+
"special": true
|
| 818 |
+
},
|
| 819 |
+
"20199": {
|
| 820 |
+
"content": "<extra_id_0>",
|
| 821 |
+
"lstrip": true,
|
| 822 |
+
"normalized": false,
|
| 823 |
+
"rstrip": true,
|
| 824 |
+
"single_word": false,
|
| 825 |
+
"special": true
|
| 826 |
+
}
|
| 827 |
+
},
|
| 828 |
"additional_special_tokens": [
|
| 829 |
"<extra_id_0>",
|
| 830 |
"<extra_id_1>",
|
trainer_state.json
CHANGED
|
@@ -1,1430 +1,1380 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 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.
|
| 13 |
-
"learning_rate":
|
| 14 |
-
"loss": 0.
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"learning_rate": 5.
|
| 20 |
-
"loss": 0.
|
| 21 |
-
"step":
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"learning_rate": 1.
|
| 26 |
-
"loss": 0.
|
| 27 |
-
"step":
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
-
"learning_rate": 1.
|
| 32 |
-
"loss": 0.
|
| 33 |
-
"step":
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"learning_rate": 2.
|
| 38 |
-
"loss": 0.
|
| 39 |
-
"step":
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"learning_rate": 2.
|
| 44 |
-
"loss": 0.
|
| 45 |
-
"step":
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
-
"learning_rate": 3.
|
| 50 |
-
"loss": 0.
|
| 51 |
-
"step":
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"learning_rate": 3.
|
| 56 |
-
"loss": 0.
|
| 57 |
-
"step":
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch":
|
| 61 |
-
"
|
| 62 |
-
"
|
| 63 |
-
"
|
| 64 |
-
"eval_runtime": 1.9415,
|
| 65 |
-
"eval_samples_per_second": 42.235,
|
| 66 |
-
"eval_steps_per_second": 1.545,
|
| 67 |
-
"step": 69
|
| 68 |
},
|
| 69 |
{
|
| 70 |
-
"epoch":
|
| 71 |
-
"learning_rate": 4.
|
| 72 |
-
"loss": 0.
|
| 73 |
-
"step":
|
| 74 |
},
|
| 75 |
{
|
| 76 |
-
"epoch":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss": 0.
|
| 79 |
-
"step":
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch":
|
| 83 |
-
"learning_rate": 5.
|
| 84 |
-
"loss": 0.
|
| 85 |
-
"step":
|
| 86 |
},
|
| 87 |
{
|
| 88 |
-
"epoch":
|
| 89 |
-
"learning_rate":
|
| 90 |
-
"loss": 0.
|
| 91 |
-
"step":
|
| 92 |
},
|
| 93 |
{
|
| 94 |
-
"epoch":
|
| 95 |
-
"learning_rate": 6.
|
| 96 |
-
"loss": 0.
|
| 97 |
-
"step":
|
| 98 |
},
|
| 99 |
{
|
| 100 |
-
"epoch": 1.
|
| 101 |
-
"
|
| 102 |
-
"
|
| 103 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 104 |
},
|
| 105 |
{
|
| 106 |
-
"epoch": 1.
|
| 107 |
-
"learning_rate": 7.
|
| 108 |
-
"loss": 0.
|
| 109 |
-
"step":
|
| 110 |
},
|
| 111 |
{
|
| 112 |
-
"epoch": 1.
|
| 113 |
-
"learning_rate": 7.
|
| 114 |
-
"loss": 0.
|
| 115 |
-
"step":
|
| 116 |
},
|
| 117 |
{
|
| 118 |
-
"epoch": 2
|
| 119 |
-
"
|
| 120 |
-
"
|
| 121 |
-
"
|
| 122 |
-
"eval_runtime": 1.9167,
|
| 123 |
-
"eval_samples_per_second": 42.781,
|
| 124 |
-
"eval_steps_per_second": 1.565,
|
| 125 |
-
"step": 138
|
| 126 |
},
|
| 127 |
{
|
| 128 |
-
"epoch":
|
| 129 |
-
"learning_rate": 8.
|
| 130 |
-
"loss": 0.
|
| 131 |
-
"step":
|
| 132 |
},
|
| 133 |
{
|
| 134 |
-
"epoch":
|
| 135 |
-
"learning_rate":
|
| 136 |
-
"loss": 0.
|
| 137 |
-
"step":
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch":
|
| 141 |
-
"learning_rate": 9.
|
| 142 |
-
"loss": 0.
|
| 143 |
-
"step":
|
| 144 |
},
|
| 145 |
{
|
| 146 |
-
"epoch":
|
| 147 |
-
"learning_rate": 9.
|
| 148 |
-
"loss": 0.
|
| 149 |
-
"step":
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch":
|
| 153 |
-
"learning_rate": 9.
|
| 154 |
-
"loss": 0.
|
| 155 |
-
"step":
|
| 156 |
},
|
| 157 |
{
|
| 158 |
-
"epoch":
|
| 159 |
-
"learning_rate": 9.
|
| 160 |
-
"loss": 0.
|
| 161 |
-
"step":
|
| 162 |
},
|
| 163 |
{
|
| 164 |
-
"epoch":
|
| 165 |
-
"learning_rate": 9.
|
| 166 |
-
"loss": 0.
|
| 167 |
-
"step":
|
| 168 |
},
|
| 169 |
{
|
| 170 |
-
"epoch":
|
| 171 |
-
"learning_rate": 9.
|
| 172 |
-
"loss": 0.
|
| 173 |
-
"step":
|
| 174 |
},
|
| 175 |
{
|
| 176 |
-
"epoch":
|
| 177 |
-
"
|
| 178 |
-
"
|
| 179 |
-
"
|
| 180 |
-
"eval_runtime": 1.9341,
|
| 181 |
-
"eval_samples_per_second": 42.396,
|
| 182 |
-
"eval_steps_per_second": 1.551,
|
| 183 |
-
"step": 207
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch":
|
| 187 |
-
"learning_rate": 9.
|
| 188 |
-
"loss": 0.
|
| 189 |
-
"step":
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch":
|
| 193 |
-
"
|
| 194 |
-
"
|
| 195 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch":
|
| 199 |
-
"learning_rate": 9.
|
| 200 |
-
"loss": 0.
|
| 201 |
-
"step":
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch":
|
| 205 |
-
"learning_rate": 9.
|
| 206 |
-
"loss": 0.
|
| 207 |
-
"step":
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch":
|
| 211 |
-
"learning_rate": 9.
|
| 212 |
-
"loss": 0.
|
| 213 |
-
"step":
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch":
|
| 217 |
-
"learning_rate": 9.
|
| 218 |
-
"loss": 0.
|
| 219 |
-
"step":
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch":
|
| 223 |
-
"learning_rate": 9.
|
| 224 |
"loss": 0.0004,
|
| 225 |
-
"step":
|
| 226 |
-
},
|
| 227 |
-
{
|
| 228 |
-
"epoch": 4.0,
|
| 229 |
-
"eval_loss": 0.0036886015441268682,
|
| 230 |
-
"eval_max_distance": 3,
|
| 231 |
-
"eval_mean_distance": 0,
|
| 232 |
-
"eval_runtime": 1.9271,
|
| 233 |
-
"eval_samples_per_second": 42.551,
|
| 234 |
-
"eval_steps_per_second": 1.557,
|
| 235 |
-
"step": 276
|
| 236 |
},
|
| 237 |
{
|
| 238 |
-
"epoch":
|
| 239 |
-
"learning_rate": 9.
|
| 240 |
-
"loss": 0.
|
| 241 |
-
"step":
|
| 242 |
},
|
| 243 |
{
|
| 244 |
-
"epoch":
|
| 245 |
-
"learning_rate": 9.
|
| 246 |
-
"loss": 0.
|
| 247 |
-
"step":
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch":
|
| 251 |
-
"learning_rate": 9.
|
| 252 |
-
"loss": 0.
|
| 253 |
-
"step":
|
| 254 |
},
|
| 255 |
{
|
| 256 |
-
"epoch":
|
| 257 |
-
"learning_rate": 9.
|
| 258 |
-
"loss": 0.
|
| 259 |
-
"step":
|
| 260 |
},
|
| 261 |
{
|
| 262 |
-
"epoch":
|
| 263 |
-
"learning_rate": 9.
|
| 264 |
-
"loss": 0.
|
| 265 |
-
"step":
|
| 266 |
},
|
| 267 |
{
|
| 268 |
-
"epoch":
|
| 269 |
-
"learning_rate": 9.
|
| 270 |
-
"loss": 0.
|
| 271 |
-
"step":
|
| 272 |
},
|
| 273 |
{
|
| 274 |
-
"epoch":
|
| 275 |
-
"learning_rate": 8.
|
| 276 |
-
"loss": 0.
|
| 277 |
-
"step":
|
| 278 |
},
|
| 279 |
{
|
| 280 |
-
"epoch":
|
| 281 |
-
"learning_rate": 8.
|
| 282 |
-
"loss": 0.
|
| 283 |
-
"step":
|
| 284 |
},
|
| 285 |
{
|
| 286 |
-
"epoch":
|
| 287 |
-
"eval_loss": 0.
|
| 288 |
-
"eval_max_distance":
|
| 289 |
"eval_mean_distance": 0,
|
| 290 |
-
"eval_runtime":
|
| 291 |
-
"eval_samples_per_second":
|
| 292 |
-
"eval_steps_per_second":
|
| 293 |
-
"step":
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch":
|
| 297 |
-
"learning_rate": 8.
|
| 298 |
-
"loss": 0.
|
| 299 |
-
"step":
|
| 300 |
},
|
| 301 |
{
|
| 302 |
-
"epoch":
|
| 303 |
-
"learning_rate": 8.
|
| 304 |
-
"loss": 0.
|
| 305 |
-
"step":
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"epoch":
|
| 309 |
-
"learning_rate": 8.
|
| 310 |
-
"loss": 0.
|
| 311 |
-
"step":
|
| 312 |
},
|
| 313 |
{
|
| 314 |
-
"epoch":
|
| 315 |
-
"learning_rate": 8.
|
| 316 |
-
"loss": 0.
|
| 317 |
-
"step":
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch":
|
| 321 |
-
"learning_rate": 8.
|
| 322 |
-
"loss": 0.
|
| 323 |
-
"step":
|
| 324 |
},
|
| 325 |
{
|
| 326 |
-
"epoch":
|
| 327 |
-
"learning_rate": 8.
|
| 328 |
-
"loss": 0.
|
| 329 |
-
"step":
|
| 330 |
},
|
| 331 |
{
|
| 332 |
-
"epoch":
|
| 333 |
-
"learning_rate": 8.
|
| 334 |
-
"loss": 0.
|
| 335 |
-
"step":
|
| 336 |
},
|
| 337 |
{
|
| 338 |
-
"epoch":
|
| 339 |
-
"learning_rate": 8.
|
| 340 |
-
"loss": 0.
|
| 341 |
-
"step":
|
| 342 |
},
|
| 343 |
{
|
| 344 |
-
"epoch":
|
| 345 |
-
"
|
| 346 |
-
"
|
| 347 |
-
"
|
| 348 |
-
"eval_runtime": 1.9399,
|
| 349 |
-
"eval_samples_per_second": 42.271,
|
| 350 |
-
"eval_steps_per_second": 1.546,
|
| 351 |
-
"step": 414
|
| 352 |
},
|
| 353 |
{
|
| 354 |
-
"epoch":
|
| 355 |
-
"learning_rate": 8.
|
| 356 |
-
"loss": 0.
|
| 357 |
-
"step":
|
| 358 |
},
|
| 359 |
{
|
| 360 |
-
"epoch":
|
| 361 |
-
"learning_rate": 8.
|
| 362 |
-
"loss": 0.
|
| 363 |
-
"step":
|
| 364 |
},
|
| 365 |
{
|
| 366 |
-
"epoch":
|
| 367 |
-
"learning_rate": 8.
|
| 368 |
-
"loss": 0.
|
| 369 |
-
"step":
|
| 370 |
},
|
| 371 |
{
|
| 372 |
-
"epoch":
|
| 373 |
-
"learning_rate": 8.
|
| 374 |
-
"loss": 0.
|
| 375 |
-
"step":
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch":
|
| 379 |
-
"learning_rate": 8.
|
| 380 |
-
"loss": 0.
|
| 381 |
-
"step":
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch":
|
| 385 |
-
"
|
| 386 |
-
"
|
| 387 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch":
|
| 391 |
-
"learning_rate": 8.
|
| 392 |
-
"loss": 0.
|
| 393 |
-
"step":
|
| 394 |
},
|
| 395 |
{
|
| 396 |
-
"epoch":
|
| 397 |
-
"
|
| 398 |
-
"
|
| 399 |
-
"
|
| 400 |
-
"eval_runtime": 1.9188,
|
| 401 |
-
"eval_samples_per_second": 42.735,
|
| 402 |
-
"eval_steps_per_second": 1.563,
|
| 403 |
-
"step": 483
|
| 404 |
},
|
| 405 |
{
|
| 406 |
-
"epoch":
|
| 407 |
-
"learning_rate": 7.
|
| 408 |
-
"loss": 0.
|
| 409 |
-
"step":
|
| 410 |
},
|
| 411 |
{
|
| 412 |
-
"epoch":
|
| 413 |
-
"learning_rate": 7.
|
| 414 |
-
"loss": 0.
|
| 415 |
-
"step":
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch":
|
| 419 |
-
"learning_rate": 7.
|
| 420 |
-
"loss": 0.
|
| 421 |
-
"step":
|
| 422 |
},
|
| 423 |
{
|
| 424 |
-
"epoch":
|
| 425 |
-
"learning_rate": 7.
|
| 426 |
-
"loss": 0.
|
| 427 |
-
"step":
|
| 428 |
},
|
| 429 |
{
|
| 430 |
-
"epoch":
|
| 431 |
-
"learning_rate": 7.
|
| 432 |
-
"loss": 0.
|
| 433 |
-
"step":
|
| 434 |
},
|
| 435 |
{
|
| 436 |
-
"epoch":
|
| 437 |
-
"learning_rate": 7.
|
| 438 |
-
"loss": 0.
|
| 439 |
-
"step":
|
| 440 |
},
|
| 441 |
{
|
| 442 |
-
"epoch":
|
| 443 |
-
"learning_rate": 7.
|
| 444 |
-
"loss": 0.
|
| 445 |
-
"step":
|
| 446 |
},
|
| 447 |
{
|
| 448 |
-
"epoch":
|
| 449 |
-
"learning_rate": 7.
|
| 450 |
-
"loss": 0.
|
| 451 |
-
"step":
|
| 452 |
},
|
| 453 |
{
|
| 454 |
-
"epoch":
|
| 455 |
-
"
|
| 456 |
-
"
|
| 457 |
-
"
|
| 458 |
-
"eval_runtime": 1.9435,
|
| 459 |
-
"eval_samples_per_second": 42.193,
|
| 460 |
-
"eval_steps_per_second": 1.544,
|
| 461 |
-
"step": 552
|
| 462 |
},
|
| 463 |
{
|
| 464 |
-
"epoch":
|
| 465 |
-
"learning_rate": 7.
|
| 466 |
-
"loss": 0.
|
| 467 |
-
"step":
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch":
|
| 471 |
-
"learning_rate": 7.
|
| 472 |
-
"loss": 0.
|
| 473 |
-
"step":
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"epoch":
|
| 477 |
-
"
|
| 478 |
-
"
|
| 479 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 480 |
},
|
| 481 |
{
|
| 482 |
-
"epoch":
|
| 483 |
-
"learning_rate": 7.
|
| 484 |
-
"loss": 0.
|
| 485 |
-
"step":
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch":
|
| 489 |
-
"learning_rate": 7.
|
| 490 |
-
"loss": 0.
|
| 491 |
-
"step":
|
| 492 |
},
|
| 493 |
{
|
| 494 |
-
"epoch":
|
| 495 |
-
"learning_rate": 7.
|
| 496 |
-
"loss": 0.
|
| 497 |
-
"step":
|
| 498 |
},
|
| 499 |
{
|
| 500 |
-
"epoch":
|
| 501 |
-
"learning_rate": 7.
|
| 502 |
-
"loss": 0.
|
| 503 |
-
"step":
|
| 504 |
},
|
| 505 |
{
|
| 506 |
-
"epoch":
|
| 507 |
-
"learning_rate": 7.
|
| 508 |
-
"loss": 0.
|
| 509 |
-
"step":
|
| 510 |
},
|
| 511 |
{
|
| 512 |
-
"epoch":
|
| 513 |
-
"
|
| 514 |
-
"
|
| 515 |
-
"
|
| 516 |
-
"eval_runtime": 1.995,
|
| 517 |
-
"eval_samples_per_second": 41.102,
|
| 518 |
-
"eval_steps_per_second": 1.504,
|
| 519 |
-
"step": 621
|
| 520 |
},
|
| 521 |
{
|
| 522 |
-
"epoch":
|
| 523 |
-
"learning_rate": 7.
|
| 524 |
-
"loss": 0.
|
| 525 |
-
"step":
|
| 526 |
},
|
| 527 |
{
|
| 528 |
-
"epoch":
|
| 529 |
-
"learning_rate": 6.
|
| 530 |
-
"loss": 0.
|
| 531 |
-
"step":
|
| 532 |
},
|
| 533 |
{
|
| 534 |
-
"epoch":
|
| 535 |
-
"learning_rate": 6.
|
| 536 |
-
"loss": 0.
|
| 537 |
-
"step":
|
| 538 |
},
|
| 539 |
{
|
| 540 |
-
"epoch":
|
| 541 |
-
"learning_rate": 6.
|
| 542 |
"loss": 0.0002,
|
| 543 |
-
"step":
|
| 544 |
},
|
| 545 |
{
|
| 546 |
-
"epoch":
|
| 547 |
-
"learning_rate": 6.
|
| 548 |
-
"loss": 0.
|
| 549 |
-
"step":
|
| 550 |
},
|
| 551 |
{
|
| 552 |
-
"epoch":
|
| 553 |
-
"learning_rate": 6.
|
| 554 |
-
"loss": 0.
|
| 555 |
-
"step":
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch":
|
| 559 |
-
"learning_rate": 6.
|
| 560 |
-
"loss": 0.
|
| 561 |
-
"step":
|
| 562 |
},
|
| 563 |
{
|
| 564 |
-
"epoch":
|
| 565 |
-
"eval_loss": 0.
|
| 566 |
-
"eval_max_distance":
|
| 567 |
"eval_mean_distance": 0,
|
| 568 |
-
"eval_runtime":
|
| 569 |
-
"eval_samples_per_second":
|
| 570 |
-
"eval_steps_per_second":
|
| 571 |
-
"step":
|
| 572 |
-
},
|
| 573 |
-
{
|
| 574 |
-
"epoch": 10.04,
|
| 575 |
-
"learning_rate": 6.649484536082475e-05,
|
| 576 |
-
"loss": 0.0011,
|
| 577 |
-
"step": 693
|
| 578 |
},
|
| 579 |
{
|
| 580 |
-
"epoch":
|
| 581 |
-
"learning_rate": 6.
|
| 582 |
-
"loss": 0.
|
| 583 |
-
"step":
|
| 584 |
},
|
| 585 |
{
|
| 586 |
-
"epoch":
|
| 587 |
-
"learning_rate": 6.
|
| 588 |
-
"loss": 0.
|
| 589 |
-
"step":
|
| 590 |
},
|
| 591 |
{
|
| 592 |
-
"epoch":
|
| 593 |
-
"learning_rate": 6.
|
| 594 |
-
"loss": 0.
|
| 595 |
-
"step":
|
| 596 |
},
|
| 597 |
{
|
| 598 |
-
"epoch":
|
| 599 |
-
"learning_rate": 6.
|
| 600 |
-
"loss": 0.
|
| 601 |
-
"step":
|
| 602 |
},
|
| 603 |
{
|
| 604 |
-
"epoch":
|
| 605 |
-
"learning_rate": 6.
|
| 606 |
-
"loss": 0.
|
| 607 |
-
"step":
|
| 608 |
},
|
| 609 |
{
|
| 610 |
-
"epoch":
|
| 611 |
-
"learning_rate": 6.
|
| 612 |
-
"loss": 0.
|
| 613 |
-
"step":
|
| 614 |
},
|
| 615 |
{
|
| 616 |
-
"epoch":
|
| 617 |
-
"learning_rate": 6.
|
| 618 |
-
"loss": 0.
|
| 619 |
-
"step":
|
| 620 |
},
|
| 621 |
{
|
| 622 |
-
"epoch":
|
| 623 |
-
"
|
| 624 |
-
"
|
| 625 |
-
"
|
| 626 |
-
"eval_runtime": 1.9385,
|
| 627 |
-
"eval_samples_per_second": 42.301,
|
| 628 |
-
"eval_steps_per_second": 1.548,
|
| 629 |
-
"step": 759
|
| 630 |
},
|
| 631 |
{
|
| 632 |
-
"epoch":
|
| 633 |
-
"learning_rate": 6.
|
| 634 |
-
"loss": 0.
|
| 635 |
-
"step":
|
| 636 |
},
|
| 637 |
{
|
| 638 |
-
"epoch":
|
| 639 |
-
"learning_rate": 6.
|
| 640 |
-
"loss": 0.
|
| 641 |
-
"step":
|
| 642 |
},
|
| 643 |
{
|
| 644 |
-
"epoch":
|
| 645 |
-
"learning_rate": 6.
|
| 646 |
-
"loss": 0.
|
| 647 |
-
"step":
|
| 648 |
},
|
| 649 |
{
|
| 650 |
-
"epoch":
|
| 651 |
-
"learning_rate": 6.
|
| 652 |
-
"loss": 0.
|
| 653 |
-
"step":
|
| 654 |
},
|
| 655 |
{
|
| 656 |
-
"epoch":
|
| 657 |
-
"learning_rate": 5.
|
| 658 |
-
"loss": 0.
|
| 659 |
-
"step":
|
| 660 |
},
|
| 661 |
{
|
| 662 |
-
"epoch":
|
| 663 |
-
"learning_rate": 5.
|
| 664 |
-
"loss": 0.
|
| 665 |
-
"step":
|
| 666 |
},
|
| 667 |
{
|
| 668 |
-
"epoch":
|
| 669 |
-
"
|
| 670 |
-
"
|
| 671 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 672 |
},
|
| 673 |
{
|
| 674 |
-
"epoch":
|
| 675 |
-
"learning_rate": 5.
|
| 676 |
-
"loss": 0.
|
| 677 |
-
"step":
|
| 678 |
},
|
| 679 |
{
|
| 680 |
-
"epoch":
|
| 681 |
-
"
|
| 682 |
-
"
|
| 683 |
-
"
|
| 684 |
-
"eval_runtime": 1.9448,
|
| 685 |
-
"eval_samples_per_second": 42.163,
|
| 686 |
-
"eval_steps_per_second": 1.543,
|
| 687 |
-
"step": 828
|
| 688 |
},
|
| 689 |
{
|
| 690 |
-
"epoch":
|
| 691 |
-
"learning_rate": 5.
|
| 692 |
-
"loss": 0.
|
| 693 |
-
"step":
|
| 694 |
},
|
| 695 |
{
|
| 696 |
-
"epoch":
|
| 697 |
-
"learning_rate": 5.
|
| 698 |
-
"loss": 0.
|
| 699 |
-
"step":
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"epoch":
|
| 703 |
-
"learning_rate": 5.
|
| 704 |
-
"loss": 0.
|
| 705 |
-
"step":
|
| 706 |
},
|
| 707 |
{
|
| 708 |
-
"epoch":
|
| 709 |
-
"learning_rate": 5.
|
| 710 |
-
"loss": 0.
|
| 711 |
-
"step":
|
| 712 |
},
|
| 713 |
{
|
| 714 |
-
"epoch":
|
| 715 |
-
"learning_rate": 5.
|
| 716 |
-
"loss": 0.
|
| 717 |
-
"step":
|
| 718 |
},
|
| 719 |
{
|
| 720 |
-
"epoch":
|
| 721 |
-
"learning_rate": 5.
|
| 722 |
-
"loss": 0.
|
| 723 |
-
"step":
|
| 724 |
},
|
| 725 |
{
|
| 726 |
-
"epoch":
|
| 727 |
-
"learning_rate": 5.
|
| 728 |
-
"loss": 0.
|
| 729 |
-
"step":
|
| 730 |
},
|
| 731 |
{
|
| 732 |
-
"epoch":
|
| 733 |
-
"
|
| 734 |
-
"
|
| 735 |
-
"
|
| 736 |
-
"eval_runtime": 1.9315,
|
| 737 |
-
"eval_samples_per_second": 42.454,
|
| 738 |
-
"eval_steps_per_second": 1.553,
|
| 739 |
-
"step": 897
|
| 740 |
},
|
| 741 |
{
|
| 742 |
-
"epoch":
|
| 743 |
-
"learning_rate": 5.
|
| 744 |
-
"loss": 0.
|
| 745 |
-
"step":
|
| 746 |
},
|
| 747 |
{
|
| 748 |
-
"epoch":
|
| 749 |
-
"learning_rate": 5.
|
| 750 |
-
"loss": 0.
|
| 751 |
-
"step":
|
| 752 |
},
|
| 753 |
{
|
| 754 |
-
"epoch":
|
| 755 |
-
"learning_rate": 5.
|
| 756 |
-
"loss": 0.
|
| 757 |
-
"step":
|
| 758 |
},
|
| 759 |
{
|
| 760 |
-
"epoch":
|
| 761 |
-
"
|
| 762 |
-
"
|
| 763 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 764 |
},
|
| 765 |
{
|
| 766 |
-
"epoch":
|
| 767 |
-
"learning_rate": 5.
|
| 768 |
-
"loss": 0.
|
| 769 |
-
"step":
|
| 770 |
},
|
| 771 |
{
|
| 772 |
-
"epoch":
|
| 773 |
-
"learning_rate": 5.
|
| 774 |
-
"loss": 0.
|
| 775 |
-
"step":
|
| 776 |
},
|
| 777 |
{
|
| 778 |
-
"epoch":
|
| 779 |
-
"learning_rate":
|
| 780 |
-
"loss": 0.
|
| 781 |
-
"step":
|
| 782 |
},
|
| 783 |
{
|
| 784 |
-
"epoch":
|
| 785 |
-
"learning_rate": 4.
|
| 786 |
-
"loss": 0.
|
| 787 |
-
"step":
|
| 788 |
},
|
| 789 |
{
|
| 790 |
-
"epoch":
|
| 791 |
-
"
|
| 792 |
-
"
|
| 793 |
-
"
|
| 794 |
-
"eval_runtime": 1.993,
|
| 795 |
-
"eval_samples_per_second": 41.144,
|
| 796 |
-
"eval_steps_per_second": 1.505,
|
| 797 |
-
"step": 966
|
| 798 |
},
|
| 799 |
{
|
| 800 |
-
"epoch":
|
| 801 |
-
"learning_rate": 4.
|
| 802 |
-
"loss": 0.
|
| 803 |
-
"step":
|
| 804 |
},
|
| 805 |
{
|
| 806 |
-
"epoch":
|
| 807 |
-
"learning_rate": 4.
|
| 808 |
-
"loss": 0.
|
| 809 |
-
"step":
|
| 810 |
},
|
| 811 |
{
|
| 812 |
-
"epoch":
|
| 813 |
-
"learning_rate": 4.
|
| 814 |
-
"loss": 0.
|
| 815 |
-
"step":
|
| 816 |
},
|
| 817 |
{
|
| 818 |
-
"epoch":
|
| 819 |
-
"learning_rate": 4.
|
| 820 |
-
"loss": 0.
|
| 821 |
-
"step":
|
| 822 |
},
|
| 823 |
{
|
| 824 |
-
"epoch":
|
| 825 |
-
"learning_rate": 4.
|
| 826 |
-
"loss": 0.
|
| 827 |
-
"step":
|
| 828 |
},
|
| 829 |
{
|
| 830 |
-
"epoch":
|
| 831 |
-
"learning_rate": 4.
|
| 832 |
-
"loss": 0.
|
| 833 |
-
"step":
|
| 834 |
},
|
| 835 |
{
|
| 836 |
-
"epoch":
|
| 837 |
-
"learning_rate": 4.
|
| 838 |
-
"loss": 0.
|
| 839 |
-
"step":
|
| 840 |
},
|
| 841 |
{
|
| 842 |
-
"epoch":
|
| 843 |
-
"learning_rate": 4.
|
| 844 |
-
"loss": 0.
|
| 845 |
-
"step":
|
| 846 |
},
|
| 847 |
{
|
| 848 |
-
"epoch":
|
| 849 |
-
"eval_loss": 0.
|
| 850 |
"eval_max_distance": 3,
|
| 851 |
"eval_mean_distance": 0,
|
| 852 |
-
"eval_runtime":
|
| 853 |
-
"eval_samples_per_second":
|
| 854 |
-
"eval_steps_per_second":
|
| 855 |
-
"step":
|
| 856 |
},
|
| 857 |
{
|
| 858 |
-
"epoch":
|
| 859 |
-
"learning_rate": 4.
|
| 860 |
-
"loss": 0.
|
| 861 |
-
"step":
|
| 862 |
},
|
| 863 |
{
|
| 864 |
-
"epoch":
|
| 865 |
-
"learning_rate": 4.
|
| 866 |
-
"loss": 0.
|
| 867 |
-
"step":
|
| 868 |
},
|
| 869 |
{
|
| 870 |
-
"epoch":
|
| 871 |
-
"learning_rate": 4.
|
| 872 |
-
"loss": 0.
|
| 873 |
-
"step":
|
| 874 |
},
|
| 875 |
{
|
| 876 |
-
"epoch":
|
| 877 |
-
"learning_rate": 4.
|
| 878 |
-
"loss": 0.
|
| 879 |
-
"step":
|
| 880 |
},
|
| 881 |
{
|
| 882 |
-
"epoch":
|
| 883 |
-
"learning_rate": 4.
|
| 884 |
-
"loss": 0.
|
| 885 |
-
"step":
|
| 886 |
},
|
| 887 |
{
|
| 888 |
-
"epoch":
|
| 889 |
-
"learning_rate": 4.
|
| 890 |
-
"loss": 0.
|
| 891 |
-
"step":
|
| 892 |
},
|
| 893 |
{
|
| 894 |
-
"epoch":
|
| 895 |
-
"learning_rate": 4.
|
| 896 |
-
"loss": 0.
|
| 897 |
-
"step":
|
| 898 |
},
|
| 899 |
{
|
| 900 |
-
"epoch":
|
| 901 |
-
"
|
| 902 |
-
"
|
| 903 |
-
"
|
| 904 |
-
"eval_runtime": 1.9244,
|
| 905 |
-
"eval_samples_per_second": 42.612,
|
| 906 |
-
"eval_steps_per_second": 1.559,
|
| 907 |
-
"step": 1104
|
| 908 |
},
|
| 909 |
{
|
| 910 |
-
"epoch":
|
| 911 |
-
"learning_rate": 3.
|
| 912 |
-
"loss": 0.
|
| 913 |
-
"step":
|
| 914 |
},
|
| 915 |
{
|
| 916 |
-
"epoch":
|
| 917 |
-
"learning_rate": 3.
|
| 918 |
"loss": 0.0003,
|
| 919 |
-
"step":
|
| 920 |
},
|
| 921 |
{
|
| 922 |
-
"epoch":
|
| 923 |
-
"learning_rate": 3.
|
| 924 |
-
"loss": 0.
|
| 925 |
-
"step":
|
| 926 |
},
|
| 927 |
{
|
| 928 |
-
"epoch":
|
| 929 |
-
"learning_rate": 3.
|
| 930 |
-
"loss": 0.
|
| 931 |
-
"step":
|
| 932 |
},
|
| 933 |
{
|
| 934 |
-
"epoch":
|
| 935 |
-
"learning_rate": 3.
|
| 936 |
-
"loss": 0.
|
| 937 |
-
"step":
|
| 938 |
},
|
| 939 |
{
|
| 940 |
-
"epoch":
|
| 941 |
-
"learning_rate": 3.
|
| 942 |
-
"loss": 0.
|
| 943 |
-
"step":
|
| 944 |
},
|
| 945 |
{
|
| 946 |
-
"epoch":
|
| 947 |
-
"
|
| 948 |
-
"
|
| 949 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 950 |
},
|
| 951 |
{
|
| 952 |
-
"epoch":
|
| 953 |
-
"learning_rate": 3.
|
| 954 |
-
"loss": 0.
|
| 955 |
-
"step":
|
| 956 |
},
|
| 957 |
{
|
| 958 |
-
"epoch":
|
| 959 |
-
"
|
| 960 |
-
"
|
| 961 |
-
"
|
| 962 |
-
"eval_runtime": 1.9136,
|
| 963 |
-
"eval_samples_per_second": 42.851,
|
| 964 |
-
"eval_steps_per_second": 1.568,
|
| 965 |
-
"step": 1173
|
| 966 |
},
|
| 967 |
{
|
| 968 |
-
"epoch":
|
| 969 |
-
"learning_rate": 3.
|
| 970 |
-
"loss": 0.
|
| 971 |
-
"step":
|
| 972 |
},
|
| 973 |
{
|
| 974 |
-
"epoch":
|
| 975 |
-
"learning_rate": 3.
|
| 976 |
-
"loss": 0.
|
| 977 |
-
"step":
|
| 978 |
},
|
| 979 |
{
|
| 980 |
-
"epoch":
|
| 981 |
-
"learning_rate": 3.
|
| 982 |
-
"loss": 0.
|
| 983 |
-
"step":
|
| 984 |
},
|
| 985 |
{
|
| 986 |
-
"epoch":
|
| 987 |
-
"learning_rate": 3.
|
| 988 |
-
"loss": 0.
|
| 989 |
-
"step":
|
| 990 |
},
|
| 991 |
{
|
| 992 |
-
"epoch":
|
| 993 |
-
"learning_rate": 3.
|
| 994 |
"loss": 0.0002,
|
| 995 |
-
"step":
|
| 996 |
},
|
| 997 |
{
|
| 998 |
-
"epoch":
|
| 999 |
-
"learning_rate": 3.
|
| 1000 |
-
"loss": 0.
|
| 1001 |
-
"step":
|
| 1002 |
},
|
| 1003 |
{
|
| 1004 |
-
"epoch":
|
| 1005 |
-
"learning_rate": 3.
|
| 1006 |
-
"loss": 0.
|
| 1007 |
-
"step":
|
| 1008 |
},
|
| 1009 |
{
|
| 1010 |
-
"epoch":
|
| 1011 |
-
"learning_rate": 3.
|
| 1012 |
-
"loss": 0.
|
| 1013 |
-
"step":
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
-
"epoch":
|
| 1017 |
-
"
|
| 1018 |
-
"
|
| 1019 |
-
"
|
| 1020 |
-
"eval_runtime": 1.9496,
|
| 1021 |
-
"eval_samples_per_second": 42.059,
|
| 1022 |
-
"eval_steps_per_second": 1.539,
|
| 1023 |
-
"step": 1242
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
-
"epoch":
|
| 1027 |
-
"learning_rate": 3.
|
| 1028 |
-
"loss": 0.
|
| 1029 |
-
"step":
|
| 1030 |
},
|
| 1031 |
{
|
| 1032 |
-
"epoch":
|
| 1033 |
-
"learning_rate": 2.
|
| 1034 |
-
"loss": 0.
|
| 1035 |
-
"step":
|
| 1036 |
},
|
| 1037 |
{
|
| 1038 |
-
"epoch":
|
| 1039 |
-
"
|
| 1040 |
-
"
|
| 1041 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1042 |
},
|
| 1043 |
{
|
| 1044 |
-
"epoch":
|
| 1045 |
-
"learning_rate": 2.
|
| 1046 |
-
"loss": 0.
|
| 1047 |
-
"step":
|
| 1048 |
},
|
| 1049 |
{
|
| 1050 |
-
"epoch":
|
| 1051 |
-
"learning_rate": 2.
|
| 1052 |
-
"loss": 0.
|
| 1053 |
-
"step":
|
| 1054 |
},
|
| 1055 |
{
|
| 1056 |
-
"epoch":
|
| 1057 |
-
"learning_rate": 2.
|
| 1058 |
-
"loss": 0.
|
| 1059 |
-
"step":
|
| 1060 |
},
|
| 1061 |
{
|
| 1062 |
-
"epoch":
|
| 1063 |
-
"learning_rate": 2.
|
| 1064 |
-
"loss": 0.
|
| 1065 |
-
"step":
|
| 1066 |
},
|
| 1067 |
{
|
| 1068 |
-
"epoch":
|
| 1069 |
-
"
|
| 1070 |
-
"
|
| 1071 |
-
"
|
| 1072 |
-
"eval_runtime": 1.9144,
|
| 1073 |
-
"eval_samples_per_second": 42.834,
|
| 1074 |
-
"eval_steps_per_second": 1.567,
|
| 1075 |
-
"step": 1311
|
| 1076 |
},
|
| 1077 |
{
|
| 1078 |
-
"epoch":
|
| 1079 |
-
"learning_rate": 2.
|
| 1080 |
-
"loss": 0.
|
| 1081 |
-
"step":
|
| 1082 |
},
|
| 1083 |
{
|
| 1084 |
-
"epoch":
|
| 1085 |
-
"learning_rate": 2.
|
| 1086 |
-
"loss": 0.
|
| 1087 |
-
"step":
|
| 1088 |
},
|
| 1089 |
{
|
| 1090 |
-
"epoch":
|
| 1091 |
-
"learning_rate": 2.
|
| 1092 |
-
"loss": 0.
|
| 1093 |
-
"step":
|
| 1094 |
},
|
| 1095 |
{
|
| 1096 |
-
"epoch":
|
| 1097 |
-
"learning_rate": 2.
|
| 1098 |
-
"loss": 0.
|
| 1099 |
-
"step":
|
| 1100 |
},
|
| 1101 |
{
|
| 1102 |
-
"epoch":
|
| 1103 |
-
"learning_rate": 2.
|
| 1104 |
-
"loss": 0.
|
| 1105 |
-
"step":
|
| 1106 |
},
|
| 1107 |
{
|
| 1108 |
-
"epoch":
|
| 1109 |
-
"learning_rate": 2.
|
| 1110 |
-
"loss": 0.
|
| 1111 |
-
"step":
|
| 1112 |
},
|
| 1113 |
{
|
| 1114 |
-
"epoch":
|
| 1115 |
-
"learning_rate": 2.
|
| 1116 |
-
"loss": 0.
|
| 1117 |
-
"step":
|
| 1118 |
},
|
| 1119 |
{
|
| 1120 |
-
"epoch":
|
| 1121 |
-
"learning_rate": 2.
|
| 1122 |
-
"loss": 0.
|
| 1123 |
-
"step":
|
| 1124 |
},
|
| 1125 |
{
|
| 1126 |
-
"epoch":
|
| 1127 |
-
"eval_loss": 0.
|
| 1128 |
"eval_max_distance": 3,
|
| 1129 |
"eval_mean_distance": 0,
|
| 1130 |
-
"eval_runtime":
|
| 1131 |
-
"eval_samples_per_second":
|
| 1132 |
-
"eval_steps_per_second":
|
| 1133 |
-
"step":
|
| 1134 |
},
|
| 1135 |
{
|
| 1136 |
-
"epoch":
|
| 1137 |
-
"learning_rate": 2.
|
| 1138 |
-
"loss": 0.
|
| 1139 |
-
"step":
|
| 1140 |
},
|
| 1141 |
{
|
| 1142 |
-
"epoch":
|
| 1143 |
-
"learning_rate": 2.
|
| 1144 |
-
"loss": 0.
|
| 1145 |
-
"step":
|
| 1146 |
},
|
| 1147 |
{
|
| 1148 |
-
"epoch":
|
| 1149 |
-
"learning_rate": 2.
|
| 1150 |
-
"loss": 0.
|
| 1151 |
-
"step":
|
| 1152 |
},
|
| 1153 |
{
|
| 1154 |
-
"epoch":
|
| 1155 |
-
"learning_rate": 2.
|
| 1156 |
-
"loss": 0.
|
| 1157 |
-
"step":
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
-
"epoch":
|
| 1161 |
-
"learning_rate": 1.
|
| 1162 |
-
"loss": 0.
|
| 1163 |
-
"step":
|
| 1164 |
},
|
| 1165 |
{
|
| 1166 |
-
"epoch":
|
| 1167 |
-
"learning_rate": 1.
|
| 1168 |
"loss": 0.0002,
|
| 1169 |
-
"step":
|
| 1170 |
},
|
| 1171 |
{
|
| 1172 |
-
"epoch":
|
| 1173 |
-
"learning_rate": 1.
|
| 1174 |
-
"loss": 0.
|
| 1175 |
-
"step":
|
| 1176 |
},
|
| 1177 |
{
|
| 1178 |
-
"epoch":
|
| 1179 |
-
"learning_rate": 1.
|
| 1180 |
-
"loss": 0.
|
| 1181 |
-
"step":
|
| 1182 |
},
|
| 1183 |
{
|
| 1184 |
-
"epoch":
|
| 1185 |
-
"
|
| 1186 |
-
"
|
| 1187 |
-
"
|
| 1188 |
-
"eval_runtime": 1.9342,
|
| 1189 |
-
"eval_samples_per_second": 42.395,
|
| 1190 |
-
"eval_steps_per_second": 1.551,
|
| 1191 |
-
"step": 1449
|
| 1192 |
},
|
| 1193 |
{
|
| 1194 |
-
"epoch":
|
| 1195 |
-
"learning_rate": 1.
|
| 1196 |
-
"loss": 0.
|
| 1197 |
-
"step":
|
| 1198 |
},
|
| 1199 |
{
|
| 1200 |
-
"epoch":
|
| 1201 |
-
"learning_rate": 1.
|
| 1202 |
-
"loss": 0.
|
| 1203 |
-
"step":
|
| 1204 |
},
|
| 1205 |
{
|
| 1206 |
-
"epoch":
|
| 1207 |
-
"learning_rate": 1.
|
| 1208 |
-
"loss": 0.
|
| 1209 |
-
"step":
|
| 1210 |
},
|
| 1211 |
{
|
| 1212 |
-
"epoch":
|
| 1213 |
-
"learning_rate": 1.
|
| 1214 |
-
"loss": 0.
|
| 1215 |
-
"step":
|
| 1216 |
},
|
| 1217 |
{
|
| 1218 |
-
"epoch":
|
| 1219 |
-
"
|
| 1220 |
-
"
|
| 1221 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1222 |
},
|
| 1223 |
{
|
| 1224 |
-
"epoch":
|
| 1225 |
-
"learning_rate": 1.
|
| 1226 |
-
"loss": 0.
|
| 1227 |
-
"step":
|
| 1228 |
},
|
| 1229 |
{
|
| 1230 |
-
"epoch":
|
| 1231 |
-
"learning_rate": 1.
|
| 1232 |
-
"loss": 0.
|
| 1233 |
-
"step":
|
| 1234 |
},
|
| 1235 |
{
|
| 1236 |
-
"epoch":
|
| 1237 |
-
"
|
| 1238 |
-
"
|
| 1239 |
-
"
|
| 1240 |
-
"eval_runtime": 1.9113,
|
| 1241 |
-
"eval_samples_per_second": 42.902,
|
| 1242 |
-
"eval_steps_per_second": 1.57,
|
| 1243 |
-
"step": 1518
|
| 1244 |
},
|
| 1245 |
{
|
| 1246 |
-
"epoch":
|
| 1247 |
-
"learning_rate": 1.
|
| 1248 |
-
"loss": 0.
|
| 1249 |
-
"step":
|
| 1250 |
},
|
| 1251 |
{
|
| 1252 |
-
"epoch":
|
| 1253 |
-
"learning_rate": 1.
|
| 1254 |
-
"loss": 0.
|
| 1255 |
-
"step":
|
| 1256 |
},
|
| 1257 |
{
|
| 1258 |
-
"epoch":
|
| 1259 |
-
"learning_rate": 1.
|
| 1260 |
"loss": 0.0002,
|
| 1261 |
-
"step":
|
| 1262 |
},
|
| 1263 |
{
|
| 1264 |
-
"epoch":
|
| 1265 |
-
"learning_rate": 1.
|
| 1266 |
-
"loss": 0.
|
| 1267 |
-
"step":
|
| 1268 |
},
|
| 1269 |
{
|
| 1270 |
-
"epoch":
|
| 1271 |
-
"learning_rate": 1.
|
| 1272 |
-
"loss": 0.
|
| 1273 |
-
"step":
|
| 1274 |
},
|
| 1275 |
{
|
| 1276 |
-
"epoch":
|
| 1277 |
-
"learning_rate": 1.
|
| 1278 |
-
"loss": 0.
|
| 1279 |
-
"step":
|
| 1280 |
},
|
| 1281 |
{
|
| 1282 |
-
"epoch":
|
| 1283 |
-
"learning_rate": 9.
|
| 1284 |
-
"loss": 0.
|
| 1285 |
-
"step":
|
| 1286 |
},
|
| 1287 |
{
|
| 1288 |
-
"epoch":
|
| 1289 |
-
"learning_rate": 9.
|
| 1290 |
-
"loss": 0.
|
| 1291 |
-
"step":
|
| 1292 |
},
|
| 1293 |
{
|
| 1294 |
-
"epoch":
|
| 1295 |
-
"
|
| 1296 |
-
"
|
| 1297 |
-
"
|
| 1298 |
-
"eval_runtime": 1.9675,
|
| 1299 |
-
"eval_samples_per_second": 41.677,
|
| 1300 |
-
"eval_steps_per_second": 1.525,
|
| 1301 |
-
"step": 1587
|
| 1302 |
},
|
| 1303 |
{
|
| 1304 |
-
"epoch":
|
| 1305 |
-
"learning_rate": 8.
|
| 1306 |
-
"loss": 0.
|
| 1307 |
-
"step":
|
| 1308 |
},
|
| 1309 |
{
|
| 1310 |
-
"epoch":
|
| 1311 |
-
"learning_rate": 7.
|
| 1312 |
-
"loss": 0.
|
| 1313 |
-
"step":
|
| 1314 |
},
|
| 1315 |
{
|
| 1316 |
-
"epoch":
|
| 1317 |
-
"
|
| 1318 |
-
"
|
| 1319 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
-
"epoch":
|
| 1323 |
-
"learning_rate": 6.
|
| 1324 |
-
"loss": 0.
|
| 1325 |
-
"step":
|
| 1326 |
},
|
| 1327 |
{
|
| 1328 |
-
"epoch":
|
| 1329 |
-
"learning_rate": 6.
|
| 1330 |
-
"loss": 0.
|
| 1331 |
-
"step":
|
| 1332 |
},
|
| 1333 |
{
|
| 1334 |
-
"epoch":
|
| 1335 |
-
"learning_rate": 5.
|
| 1336 |
-
"loss": 0.
|
| 1337 |
-
"step":
|
| 1338 |
},
|
| 1339 |
{
|
| 1340 |
-
"epoch":
|
| 1341 |
-
"learning_rate": 5.
|
| 1342 |
-
"loss": 0.
|
| 1343 |
-
"step":
|
| 1344 |
},
|
| 1345 |
{
|
| 1346 |
-
"epoch":
|
| 1347 |
-
"learning_rate": 4.
|
| 1348 |
-
"loss": 0.
|
| 1349 |
-
"step":
|
| 1350 |
},
|
| 1351 |
{
|
| 1352 |
-
"epoch":
|
| 1353 |
-
"
|
| 1354 |
-
"
|
| 1355 |
-
"
|
| 1356 |
-
"eval_runtime": 1.9278,
|
| 1357 |
-
"eval_samples_per_second": 42.536,
|
| 1358 |
-
"eval_steps_per_second": 1.556,
|
| 1359 |
-
"step": 1656
|
| 1360 |
},
|
| 1361 |
{
|
| 1362 |
-
"epoch":
|
| 1363 |
-
"learning_rate": 3.
|
| 1364 |
-
"loss": 0.
|
| 1365 |
-
"step":
|
| 1366 |
},
|
| 1367 |
{
|
| 1368 |
-
"epoch":
|
| 1369 |
-
"learning_rate": 3.
|
| 1370 |
-
"loss": 0.
|
| 1371 |
-
"step":
|
| 1372 |
},
|
| 1373 |
{
|
| 1374 |
-
"epoch":
|
| 1375 |
-
"learning_rate": 2.
|
| 1376 |
-
"loss": 0.
|
| 1377 |
-
"step":
|
| 1378 |
},
|
| 1379 |
{
|
| 1380 |
-
"epoch":
|
| 1381 |
-
"learning_rate":
|
| 1382 |
-
"loss": 0.
|
| 1383 |
-
"step":
|
| 1384 |
},
|
| 1385 |
{
|
| 1386 |
-
"epoch":
|
| 1387 |
-
"learning_rate": 1.
|
| 1388 |
-
"loss": 0.
|
| 1389 |
-
"step":
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
-
"epoch":
|
| 1393 |
-
"learning_rate":
|
| 1394 |
-
"loss": 0.
|
| 1395 |
-
"step":
|
| 1396 |
},
|
| 1397 |
{
|
| 1398 |
-
"epoch":
|
| 1399 |
-
"learning_rate":
|
| 1400 |
-
"loss": 0.
|
| 1401 |
-
"step":
|
| 1402 |
},
|
| 1403 |
{
|
| 1404 |
-
"epoch":
|
| 1405 |
-
"eval_loss": 0.
|
| 1406 |
-
"eval_max_distance":
|
| 1407 |
"eval_mean_distance": 0,
|
| 1408 |
-
"eval_runtime":
|
| 1409 |
-
"eval_samples_per_second":
|
| 1410 |
-
"eval_steps_per_second":
|
| 1411 |
-
"step":
|
| 1412 |
-
},
|
| 1413 |
-
{
|
| 1414 |
-
"epoch":
|
| 1415 |
-
"step":
|
| 1416 |
-
"total_flos":
|
| 1417 |
-
"train_loss": 0.
|
| 1418 |
-
"train_runtime":
|
| 1419 |
-
"train_samples_per_second":
|
| 1420 |
-
"train_steps_per_second":
|
| 1421 |
}
|
| 1422 |
],
|
| 1423 |
-
"logging_steps":
|
| 1424 |
-
"max_steps":
|
| 1425 |
-
"
|
| 1426 |
-
"
|
| 1427 |
-
"
|
|
|
|
|
|
|
| 1428 |
"trial_name": null,
|
| 1429 |
"trial_params": null
|
| 1430 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 15.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 45885,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.0,
|
| 13 |
+
"learning_rate": 2.1791239921551537e-08,
|
| 14 |
+
"loss": 0.0002,
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.08,
|
| 19 |
+
"learning_rate": 5.011985181956853e-06,
|
| 20 |
+
"loss": 0.0017,
|
| 21 |
+
"step": 230
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.15,
|
| 25 |
+
"learning_rate": 1.0023970363913707e-05,
|
| 26 |
+
"loss": 0.0012,
|
| 27 |
+
"step": 460
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.23,
|
| 31 |
+
"learning_rate": 1.503595554587056e-05,
|
| 32 |
+
"loss": 0.0008,
|
| 33 |
+
"step": 690
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.3,
|
| 37 |
+
"learning_rate": 2.0047940727827413e-05,
|
| 38 |
+
"loss": 0.001,
|
| 39 |
+
"step": 920
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 0.38,
|
| 43 |
+
"learning_rate": 2.5059925909784264e-05,
|
| 44 |
+
"loss": 0.0008,
|
| 45 |
+
"step": 1150
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.45,
|
| 49 |
+
"learning_rate": 3.007191109174112e-05,
|
| 50 |
+
"loss": 0.0007,
|
| 51 |
+
"step": 1380
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.53,
|
| 55 |
+
"learning_rate": 3.5083896273697975e-05,
|
| 56 |
+
"loss": 0.0007,
|
| 57 |
+
"step": 1610
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 0.6,
|
| 61 |
+
"learning_rate": 4.0095881455654826e-05,
|
| 62 |
+
"loss": 0.0006,
|
| 63 |
+
"step": 1840
|
|
|
|
|
|
|
|
|
|
|
|
|
| 64 |
},
|
| 65 |
{
|
| 66 |
+
"epoch": 0.68,
|
| 67 |
+
"learning_rate": 4.5107866637611684e-05,
|
| 68 |
+
"loss": 0.0006,
|
| 69 |
+
"step": 2070
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 0.75,
|
| 73 |
+
"learning_rate": 5.011985181956853e-05,
|
| 74 |
+
"loss": 0.0005,
|
| 75 |
+
"step": 2300
|
| 76 |
},
|
| 77 |
{
|
| 78 |
+
"epoch": 0.83,
|
| 79 |
+
"learning_rate": 5.5131837001525385e-05,
|
| 80 |
+
"loss": 0.0004,
|
| 81 |
+
"step": 2530
|
| 82 |
},
|
| 83 |
{
|
| 84 |
+
"epoch": 0.9,
|
| 85 |
+
"learning_rate": 6.014382218348224e-05,
|
| 86 |
+
"loss": 0.0003,
|
| 87 |
+
"step": 2760
|
| 88 |
},
|
| 89 |
{
|
| 90 |
+
"epoch": 0.98,
|
| 91 |
+
"learning_rate": 6.51558073654391e-05,
|
| 92 |
+
"loss": 0.0004,
|
| 93 |
+
"step": 2990
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 1.0,
|
| 97 |
+
"eval_loss": 0.002156034577637911,
|
| 98 |
+
"eval_max_distance": 12,
|
| 99 |
+
"eval_mean_distance": 0,
|
| 100 |
+
"eval_runtime": 31.367,
|
| 101 |
+
"eval_samples_per_second": 18.778,
|
| 102 |
+
"eval_steps_per_second": 0.128,
|
| 103 |
+
"step": 3059
|
| 104 |
},
|
| 105 |
{
|
| 106 |
+
"epoch": 1.05,
|
| 107 |
+
"learning_rate": 7.016779254739595e-05,
|
| 108 |
+
"loss": 0.0004,
|
| 109 |
+
"step": 3220
|
| 110 |
},
|
| 111 |
{
|
| 112 |
+
"epoch": 1.13,
|
| 113 |
+
"learning_rate": 7.517977772935281e-05,
|
| 114 |
+
"loss": 0.0003,
|
| 115 |
+
"step": 3450
|
| 116 |
},
|
| 117 |
{
|
| 118 |
+
"epoch": 1.2,
|
| 119 |
+
"learning_rate": 8.019176291130965e-05,
|
| 120 |
+
"loss": 0.0004,
|
| 121 |
+
"step": 3680
|
|
|
|
|
|
|
|
|
|
|
|
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 1.28,
|
| 125 |
+
"learning_rate": 8.520374809326651e-05,
|
| 126 |
+
"loss": 0.0003,
|
| 127 |
+
"step": 3910
|
| 128 |
},
|
| 129 |
{
|
| 130 |
+
"epoch": 1.35,
|
| 131 |
+
"learning_rate": 9.021573327522337e-05,
|
| 132 |
+
"loss": 0.0004,
|
| 133 |
+
"step": 4140
|
| 134 |
},
|
| 135 |
{
|
| 136 |
+
"epoch": 1.43,
|
| 137 |
+
"learning_rate": 9.522771845718021e-05,
|
| 138 |
+
"loss": 0.0003,
|
| 139 |
+
"step": 4370
|
| 140 |
},
|
| 141 |
{
|
| 142 |
+
"epoch": 1.5,
|
| 143 |
+
"learning_rate": 9.997336303758234e-05,
|
| 144 |
+
"loss": 0.0003,
|
| 145 |
+
"step": 4600
|
| 146 |
},
|
| 147 |
{
|
| 148 |
+
"epoch": 1.58,
|
| 149 |
+
"learning_rate": 9.941640836884929e-05,
|
| 150 |
+
"loss": 0.0003,
|
| 151 |
+
"step": 4830
|
| 152 |
},
|
| 153 |
{
|
| 154 |
+
"epoch": 1.65,
|
| 155 |
+
"learning_rate": 9.885945370011624e-05,
|
| 156 |
+
"loss": 0.0004,
|
| 157 |
+
"step": 5060
|
| 158 |
},
|
| 159 |
{
|
| 160 |
+
"epoch": 1.73,
|
| 161 |
+
"learning_rate": 9.830249903138319e-05,
|
| 162 |
+
"loss": 0.0003,
|
| 163 |
+
"step": 5290
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 1.8,
|
| 167 |
+
"learning_rate": 9.774554436265014e-05,
|
| 168 |
+
"loss": 0.0004,
|
| 169 |
+
"step": 5520
|
| 170 |
},
|
| 171 |
{
|
| 172 |
+
"epoch": 1.88,
|
| 173 |
+
"learning_rate": 9.718858969391709e-05,
|
| 174 |
+
"loss": 0.0003,
|
| 175 |
+
"step": 5750
|
|
|
|
|
|
|
|
|
|
|
|
|
| 176 |
},
|
| 177 |
{
|
| 178 |
+
"epoch": 1.95,
|
| 179 |
+
"learning_rate": 9.663163502518404e-05,
|
| 180 |
+
"loss": 0.0003,
|
| 181 |
+
"step": 5980
|
| 182 |
},
|
| 183 |
{
|
| 184 |
+
"epoch": 2.0,
|
| 185 |
+
"eval_loss": 0.0011327904649078846,
|
| 186 |
+
"eval_max_distance": 5,
|
| 187 |
+
"eval_mean_distance": 0,
|
| 188 |
+
"eval_runtime": 30.6053,
|
| 189 |
+
"eval_samples_per_second": 19.245,
|
| 190 |
+
"eval_steps_per_second": 0.131,
|
| 191 |
+
"step": 6118
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 2.03,
|
| 195 |
+
"learning_rate": 9.6074680356451e-05,
|
| 196 |
+
"loss": 0.0003,
|
| 197 |
+
"step": 6210
|
| 198 |
},
|
| 199 |
{
|
| 200 |
+
"epoch": 2.11,
|
| 201 |
+
"learning_rate": 9.551772568771794e-05,
|
| 202 |
+
"loss": 0.0003,
|
| 203 |
+
"step": 6440
|
| 204 |
},
|
| 205 |
{
|
| 206 |
+
"epoch": 2.18,
|
| 207 |
+
"learning_rate": 9.496077101898489e-05,
|
| 208 |
+
"loss": 0.0003,
|
| 209 |
+
"step": 6670
|
| 210 |
},
|
| 211 |
{
|
| 212 |
+
"epoch": 2.26,
|
| 213 |
+
"learning_rate": 9.440381635025185e-05,
|
| 214 |
+
"loss": 0.0003,
|
| 215 |
+
"step": 6900
|
| 216 |
},
|
| 217 |
{
|
| 218 |
+
"epoch": 2.33,
|
| 219 |
+
"learning_rate": 9.384686168151879e-05,
|
| 220 |
"loss": 0.0004,
|
| 221 |
+
"step": 7130
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 222 |
},
|
| 223 |
{
|
| 224 |
+
"epoch": 2.41,
|
| 225 |
+
"learning_rate": 9.328990701278575e-05,
|
| 226 |
+
"loss": 0.0003,
|
| 227 |
+
"step": 7360
|
| 228 |
},
|
| 229 |
{
|
| 230 |
+
"epoch": 2.48,
|
| 231 |
+
"learning_rate": 9.273295234405269e-05,
|
| 232 |
+
"loss": 0.0003,
|
| 233 |
+
"step": 7590
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 2.56,
|
| 237 |
+
"learning_rate": 9.217599767531964e-05,
|
| 238 |
+
"loss": 0.0003,
|
| 239 |
+
"step": 7820
|
| 240 |
},
|
| 241 |
{
|
| 242 |
+
"epoch": 2.63,
|
| 243 |
+
"learning_rate": 9.16190430065866e-05,
|
| 244 |
+
"loss": 0.0003,
|
| 245 |
+
"step": 8050
|
| 246 |
},
|
| 247 |
{
|
| 248 |
+
"epoch": 2.71,
|
| 249 |
+
"learning_rate": 9.106208833785354e-05,
|
| 250 |
+
"loss": 0.0003,
|
| 251 |
+
"step": 8280
|
| 252 |
},
|
| 253 |
{
|
| 254 |
+
"epoch": 2.78,
|
| 255 |
+
"learning_rate": 9.05051336691205e-05,
|
| 256 |
+
"loss": 0.0003,
|
| 257 |
+
"step": 8510
|
| 258 |
},
|
| 259 |
{
|
| 260 |
+
"epoch": 2.86,
|
| 261 |
+
"learning_rate": 8.994817900038746e-05,
|
| 262 |
+
"loss": 0.0003,
|
| 263 |
+
"step": 8740
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 2.93,
|
| 267 |
+
"learning_rate": 8.939122433165439e-05,
|
| 268 |
+
"loss": 0.0003,
|
| 269 |
+
"step": 8970
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 3.0,
|
| 273 |
+
"eval_loss": 0.0010147562716156244,
|
| 274 |
+
"eval_max_distance": 6,
|
| 275 |
"eval_mean_distance": 0,
|
| 276 |
+
"eval_runtime": 30.7374,
|
| 277 |
+
"eval_samples_per_second": 19.162,
|
| 278 |
+
"eval_steps_per_second": 0.13,
|
| 279 |
+
"step": 9177
|
| 280 |
},
|
| 281 |
{
|
| 282 |
+
"epoch": 3.01,
|
| 283 |
+
"learning_rate": 8.883426966292136e-05,
|
| 284 |
+
"loss": 0.0003,
|
| 285 |
+
"step": 9200
|
| 286 |
},
|
| 287 |
{
|
| 288 |
+
"epoch": 3.08,
|
| 289 |
+
"learning_rate": 8.82773149941883e-05,
|
| 290 |
+
"loss": 0.0002,
|
| 291 |
+
"step": 9430
|
| 292 |
},
|
| 293 |
{
|
| 294 |
+
"epoch": 3.16,
|
| 295 |
+
"learning_rate": 8.772036032545526e-05,
|
| 296 |
+
"loss": 0.0003,
|
| 297 |
+
"step": 9660
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 3.23,
|
| 301 |
+
"learning_rate": 8.716340565672221e-05,
|
| 302 |
+
"loss": 0.0002,
|
| 303 |
+
"step": 9890
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 3.31,
|
| 307 |
+
"learning_rate": 8.660645098798916e-05,
|
| 308 |
+
"loss": 0.0003,
|
| 309 |
+
"step": 10120
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 3.38,
|
| 313 |
+
"learning_rate": 8.604949631925611e-05,
|
| 314 |
+
"loss": 0.0003,
|
| 315 |
+
"step": 10350
|
| 316 |
},
|
| 317 |
{
|
| 318 |
+
"epoch": 3.46,
|
| 319 |
+
"learning_rate": 8.549254165052306e-05,
|
| 320 |
+
"loss": 0.0003,
|
| 321 |
+
"step": 10580
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 3.53,
|
| 325 |
+
"learning_rate": 8.493558698179001e-05,
|
| 326 |
+
"loss": 0.0003,
|
| 327 |
+
"step": 10810
|
| 328 |
},
|
| 329 |
{
|
| 330 |
+
"epoch": 3.61,
|
| 331 |
+
"learning_rate": 8.437863231305696e-05,
|
| 332 |
+
"loss": 0.0003,
|
| 333 |
+
"step": 11040
|
|
|
|
|
|
|
|
|
|
|
|
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 3.68,
|
| 337 |
+
"learning_rate": 8.382167764432391e-05,
|
| 338 |
+
"loss": 0.0003,
|
| 339 |
+
"step": 11270
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 3.76,
|
| 343 |
+
"learning_rate": 8.326472297559086e-05,
|
| 344 |
+
"loss": 0.0003,
|
| 345 |
+
"step": 11500
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 3.83,
|
| 349 |
+
"learning_rate": 8.270776830685781e-05,
|
| 350 |
+
"loss": 0.0003,
|
| 351 |
+
"step": 11730
|
| 352 |
},
|
| 353 |
{
|
| 354 |
+
"epoch": 3.91,
|
| 355 |
+
"learning_rate": 8.215081363812476e-05,
|
| 356 |
+
"loss": 0.0003,
|
| 357 |
+
"step": 11960
|
| 358 |
},
|
| 359 |
{
|
| 360 |
+
"epoch": 3.98,
|
| 361 |
+
"learning_rate": 8.159385896939171e-05,
|
| 362 |
+
"loss": 0.0003,
|
| 363 |
+
"step": 12190
|
| 364 |
},
|
| 365 |
{
|
| 366 |
+
"epoch": 4.0,
|
| 367 |
+
"eval_loss": 0.0011835404438897967,
|
| 368 |
+
"eval_max_distance": 3,
|
| 369 |
+
"eval_mean_distance": 0,
|
| 370 |
+
"eval_runtime": 30.6791,
|
| 371 |
+
"eval_samples_per_second": 19.199,
|
| 372 |
+
"eval_steps_per_second": 0.13,
|
| 373 |
+
"step": 12236
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 4.06,
|
| 377 |
+
"learning_rate": 8.103690430065866e-05,
|
| 378 |
+
"loss": 0.0003,
|
| 379 |
+
"step": 12420
|
| 380 |
},
|
| 381 |
{
|
| 382 |
+
"epoch": 4.14,
|
| 383 |
+
"learning_rate": 8.047994963192561e-05,
|
| 384 |
+
"loss": 0.0003,
|
| 385 |
+
"step": 12650
|
|
|
|
|
|
|
|
|
|
|
|
|
| 386 |
},
|
| 387 |
{
|
| 388 |
+
"epoch": 4.21,
|
| 389 |
+
"learning_rate": 7.992299496319256e-05,
|
| 390 |
+
"loss": 0.0003,
|
| 391 |
+
"step": 12880
|
| 392 |
},
|
| 393 |
{
|
| 394 |
+
"epoch": 4.29,
|
| 395 |
+
"learning_rate": 7.936604029445953e-05,
|
| 396 |
+
"loss": 0.0002,
|
| 397 |
+
"step": 13110
|
| 398 |
},
|
| 399 |
{
|
| 400 |
+
"epoch": 4.36,
|
| 401 |
+
"learning_rate": 7.880908562572646e-05,
|
| 402 |
+
"loss": 0.0003,
|
| 403 |
+
"step": 13340
|
| 404 |
},
|
| 405 |
{
|
| 406 |
+
"epoch": 4.44,
|
| 407 |
+
"learning_rate": 7.825213095699341e-05,
|
| 408 |
+
"loss": 0.0003,
|
| 409 |
+
"step": 13570
|
| 410 |
},
|
| 411 |
{
|
| 412 |
+
"epoch": 4.51,
|
| 413 |
+
"learning_rate": 7.769517628826036e-05,
|
| 414 |
+
"loss": 0.0003,
|
| 415 |
+
"step": 13800
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 4.59,
|
| 419 |
+
"learning_rate": 7.713822161952731e-05,
|
| 420 |
+
"loss": 0.0003,
|
| 421 |
+
"step": 14030
|
| 422 |
},
|
| 423 |
{
|
| 424 |
+
"epoch": 4.66,
|
| 425 |
+
"learning_rate": 7.658126695079428e-05,
|
| 426 |
+
"loss": 0.0002,
|
| 427 |
+
"step": 14260
|
| 428 |
},
|
| 429 |
{
|
| 430 |
+
"epoch": 4.74,
|
| 431 |
+
"learning_rate": 7.602431228206122e-05,
|
| 432 |
+
"loss": 0.0003,
|
| 433 |
+
"step": 14490
|
| 434 |
},
|
| 435 |
{
|
| 436 |
+
"epoch": 4.81,
|
| 437 |
+
"learning_rate": 7.546735761332817e-05,
|
| 438 |
+
"loss": 0.0003,
|
| 439 |
+
"step": 14720
|
|
|
|
|
|
|
|
|
|
|
|
|
| 440 |
},
|
| 441 |
{
|
| 442 |
+
"epoch": 4.89,
|
| 443 |
+
"learning_rate": 7.491040294459513e-05,
|
| 444 |
+
"loss": 0.0002,
|
| 445 |
+
"step": 14950
|
| 446 |
},
|
| 447 |
{
|
| 448 |
+
"epoch": 4.96,
|
| 449 |
+
"learning_rate": 7.435344827586207e-05,
|
| 450 |
+
"loss": 0.0003,
|
| 451 |
+
"step": 15180
|
| 452 |
},
|
| 453 |
{
|
| 454 |
+
"epoch": 5.0,
|
| 455 |
+
"eval_loss": 0.000761075527407229,
|
| 456 |
+
"eval_max_distance": 3,
|
| 457 |
+
"eval_mean_distance": 0,
|
| 458 |
+
"eval_runtime": 30.6946,
|
| 459 |
+
"eval_samples_per_second": 19.189,
|
| 460 |
+
"eval_steps_per_second": 0.13,
|
| 461 |
+
"step": 15295
|
| 462 |
},
|
| 463 |
{
|
| 464 |
+
"epoch": 5.04,
|
| 465 |
+
"learning_rate": 7.379649360712903e-05,
|
| 466 |
+
"loss": 0.0002,
|
| 467 |
+
"step": 15410
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 5.11,
|
| 471 |
+
"learning_rate": 7.323953893839598e-05,
|
| 472 |
+
"loss": 0.0002,
|
| 473 |
+
"step": 15640
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"epoch": 5.19,
|
| 477 |
+
"learning_rate": 7.268258426966292e-05,
|
| 478 |
+
"loss": 0.0003,
|
| 479 |
+
"step": 15870
|
| 480 |
},
|
| 481 |
{
|
| 482 |
+
"epoch": 5.26,
|
| 483 |
+
"learning_rate": 7.212562960092988e-05,
|
| 484 |
+
"loss": 0.0002,
|
| 485 |
+
"step": 16100
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 5.34,
|
| 489 |
+
"learning_rate": 7.156867493219682e-05,
|
| 490 |
+
"loss": 0.0003,
|
| 491 |
+
"step": 16330
|
| 492 |
},
|
| 493 |
{
|
| 494 |
+
"epoch": 5.41,
|
| 495 |
+
"learning_rate": 7.101172026346378e-05,
|
| 496 |
+
"loss": 0.0002,
|
| 497 |
+
"step": 16560
|
|
|
|
|
|
|
|
|
|
|
|
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 5.49,
|
| 501 |
+
"learning_rate": 7.045476559473073e-05,
|
| 502 |
+
"loss": 0.0002,
|
| 503 |
+
"step": 16790
|
| 504 |
},
|
| 505 |
{
|
| 506 |
+
"epoch": 5.56,
|
| 507 |
+
"learning_rate": 6.989781092599767e-05,
|
| 508 |
+
"loss": 0.0003,
|
| 509 |
+
"step": 17020
|
| 510 |
},
|
| 511 |
{
|
| 512 |
+
"epoch": 5.64,
|
| 513 |
+
"learning_rate": 6.934085625726463e-05,
|
| 514 |
+
"loss": 0.0003,
|
| 515 |
+
"step": 17250
|
| 516 |
},
|
| 517 |
{
|
| 518 |
+
"epoch": 5.71,
|
| 519 |
+
"learning_rate": 6.878390158853158e-05,
|
| 520 |
"loss": 0.0002,
|
| 521 |
+
"step": 17480
|
| 522 |
},
|
| 523 |
{
|
| 524 |
+
"epoch": 5.79,
|
| 525 |
+
"learning_rate": 6.822694691979853e-05,
|
| 526 |
+
"loss": 0.0003,
|
| 527 |
+
"step": 17710
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 5.86,
|
| 531 |
+
"learning_rate": 6.766999225106548e-05,
|
| 532 |
+
"loss": 0.0003,
|
| 533 |
+
"step": 17940
|
| 534 |
},
|
| 535 |
{
|
| 536 |
+
"epoch": 5.94,
|
| 537 |
+
"learning_rate": 6.711303758233242e-05,
|
| 538 |
+
"loss": 0.0002,
|
| 539 |
+
"step": 18170
|
| 540 |
},
|
| 541 |
{
|
| 542 |
+
"epoch": 6.0,
|
| 543 |
+
"eval_loss": 0.0009165782830677927,
|
| 544 |
+
"eval_max_distance": 3,
|
| 545 |
"eval_mean_distance": 0,
|
| 546 |
+
"eval_runtime": 30.7001,
|
| 547 |
+
"eval_samples_per_second": 19.186,
|
| 548 |
+
"eval_steps_per_second": 0.13,
|
| 549 |
+
"step": 18354
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 550 |
},
|
| 551 |
{
|
| 552 |
+
"epoch": 6.02,
|
| 553 |
+
"learning_rate": 6.655608291359938e-05,
|
| 554 |
+
"loss": 0.0003,
|
| 555 |
+
"step": 18400
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 6.09,
|
| 559 |
+
"learning_rate": 6.599912824486634e-05,
|
| 560 |
+
"loss": 0.0002,
|
| 561 |
+
"step": 18630
|
| 562 |
},
|
| 563 |
{
|
| 564 |
+
"epoch": 6.17,
|
| 565 |
+
"learning_rate": 6.544217357613329e-05,
|
| 566 |
+
"loss": 0.0002,
|
| 567 |
+
"step": 18860
|
| 568 |
},
|
| 569 |
{
|
| 570 |
+
"epoch": 6.24,
|
| 571 |
+
"learning_rate": 6.488521890740024e-05,
|
| 572 |
+
"loss": 0.0003,
|
| 573 |
+
"step": 19090
|
| 574 |
},
|
| 575 |
{
|
| 576 |
+
"epoch": 6.32,
|
| 577 |
+
"learning_rate": 6.432826423866719e-05,
|
| 578 |
+
"loss": 0.0002,
|
| 579 |
+
"step": 19320
|
| 580 |
},
|
| 581 |
{
|
| 582 |
+
"epoch": 6.39,
|
| 583 |
+
"learning_rate": 6.377130956993414e-05,
|
| 584 |
+
"loss": 0.0002,
|
| 585 |
+
"step": 19550
|
| 586 |
},
|
| 587 |
{
|
| 588 |
+
"epoch": 6.47,
|
| 589 |
+
"learning_rate": 6.321435490120109e-05,
|
| 590 |
+
"loss": 0.0003,
|
| 591 |
+
"step": 19780
|
| 592 |
},
|
| 593 |
{
|
| 594 |
+
"epoch": 6.54,
|
| 595 |
+
"learning_rate": 6.265740023246804e-05,
|
| 596 |
+
"loss": 0.0002,
|
| 597 |
+
"step": 20010
|
|
|
|
|
|
|
|
|
|
|
|
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 6.62,
|
| 601 |
+
"learning_rate": 6.210044556373499e-05,
|
| 602 |
+
"loss": 0.0002,
|
| 603 |
+
"step": 20240
|
| 604 |
},
|
| 605 |
{
|
| 606 |
+
"epoch": 6.69,
|
| 607 |
+
"learning_rate": 6.154349089500194e-05,
|
| 608 |
+
"loss": 0.0003,
|
| 609 |
+
"step": 20470
|
| 610 |
},
|
| 611 |
{
|
| 612 |
+
"epoch": 6.77,
|
| 613 |
+
"learning_rate": 6.098653622626889e-05,
|
| 614 |
+
"loss": 0.0002,
|
| 615 |
+
"step": 20700
|
| 616 |
},
|
| 617 |
{
|
| 618 |
+
"epoch": 6.84,
|
| 619 |
+
"learning_rate": 6.042958155753584e-05,
|
| 620 |
+
"loss": 0.0002,
|
| 621 |
+
"step": 20930
|
| 622 |
},
|
| 623 |
{
|
| 624 |
+
"epoch": 6.92,
|
| 625 |
+
"learning_rate": 5.9872626888802796e-05,
|
| 626 |
+
"loss": 0.0003,
|
| 627 |
+
"step": 21160
|
| 628 |
},
|
| 629 |
{
|
| 630 |
+
"epoch": 6.99,
|
| 631 |
+
"learning_rate": 5.931567222006974e-05,
|
| 632 |
+
"loss": 0.0002,
|
| 633 |
+
"step": 21390
|
| 634 |
},
|
| 635 |
{
|
| 636 |
+
"epoch": 7.0,
|
| 637 |
+
"eval_loss": 0.0007532148738391697,
|
| 638 |
+
"eval_max_distance": 3,
|
| 639 |
+
"eval_mean_distance": 0,
|
| 640 |
+
"eval_runtime": 31.1332,
|
| 641 |
+
"eval_samples_per_second": 18.919,
|
| 642 |
+
"eval_steps_per_second": 0.128,
|
| 643 |
+
"step": 21413
|
| 644 |
},
|
| 645 |
{
|
| 646 |
+
"epoch": 7.07,
|
| 647 |
+
"learning_rate": 5.87587175513367e-05,
|
| 648 |
+
"loss": 0.0003,
|
| 649 |
+
"step": 21620
|
| 650 |
},
|
| 651 |
{
|
| 652 |
+
"epoch": 7.14,
|
| 653 |
+
"learning_rate": 5.820176288260365e-05,
|
| 654 |
+
"loss": 0.0002,
|
| 655 |
+
"step": 21850
|
|
|
|
|
|
|
|
|
|
|
|
|
| 656 |
},
|
| 657 |
{
|
| 658 |
+
"epoch": 7.22,
|
| 659 |
+
"learning_rate": 5.764480821387059e-05,
|
| 660 |
+
"loss": 0.0002,
|
| 661 |
+
"step": 22080
|
| 662 |
},
|
| 663 |
{
|
| 664 |
+
"epoch": 7.29,
|
| 665 |
+
"learning_rate": 5.708785354513755e-05,
|
| 666 |
+
"loss": 0.0002,
|
| 667 |
+
"step": 22310
|
| 668 |
},
|
| 669 |
{
|
| 670 |
+
"epoch": 7.37,
|
| 671 |
+
"learning_rate": 5.653089887640449e-05,
|
| 672 |
+
"loss": 0.0003,
|
| 673 |
+
"step": 22540
|
| 674 |
},
|
| 675 |
{
|
| 676 |
+
"epoch": 7.44,
|
| 677 |
+
"learning_rate": 5.597394420767145e-05,
|
| 678 |
+
"loss": 0.0003,
|
| 679 |
+
"step": 22770
|
| 680 |
},
|
| 681 |
{
|
| 682 |
+
"epoch": 7.52,
|
| 683 |
+
"learning_rate": 5.54169895389384e-05,
|
| 684 |
+
"loss": 0.0002,
|
| 685 |
+
"step": 23000
|
| 686 |
},
|
| 687 |
{
|
| 688 |
+
"epoch": 7.59,
|
| 689 |
+
"learning_rate": 5.486003487020534e-05,
|
| 690 |
+
"loss": 0.0002,
|
| 691 |
+
"step": 23230
|
| 692 |
},
|
| 693 |
{
|
| 694 |
+
"epoch": 7.67,
|
| 695 |
+
"learning_rate": 5.43030802014723e-05,
|
| 696 |
+
"loss": 0.0003,
|
| 697 |
+
"step": 23460
|
| 698 |
},
|
| 699 |
{
|
| 700 |
+
"epoch": 7.74,
|
| 701 |
+
"learning_rate": 5.374612553273926e-05,
|
| 702 |
+
"loss": 0.0002,
|
| 703 |
+
"step": 23690
|
|
|
|
|
|
|
|
|
|
|
|
|
| 704 |
},
|
| 705 |
{
|
| 706 |
+
"epoch": 7.82,
|
| 707 |
+
"learning_rate": 5.31891708640062e-05,
|
| 708 |
+
"loss": 0.0002,
|
| 709 |
+
"step": 23920
|
| 710 |
},
|
| 711 |
{
|
| 712 |
+
"epoch": 7.89,
|
| 713 |
+
"learning_rate": 5.263221619527316e-05,
|
| 714 |
+
"loss": 0.0002,
|
| 715 |
+
"step": 24150
|
| 716 |
},
|
| 717 |
{
|
| 718 |
+
"epoch": 7.97,
|
| 719 |
+
"learning_rate": 5.20752615265401e-05,
|
| 720 |
+
"loss": 0.0002,
|
| 721 |
+
"step": 24380
|
| 722 |
},
|
| 723 |
{
|
| 724 |
+
"epoch": 8.0,
|
| 725 |
+
"eval_loss": 0.0007926285616122186,
|
| 726 |
+
"eval_max_distance": 3,
|
| 727 |
+
"eval_mean_distance": 0,
|
| 728 |
+
"eval_runtime": 30.7407,
|
| 729 |
+
"eval_samples_per_second": 19.16,
|
| 730 |
+
"eval_steps_per_second": 0.13,
|
| 731 |
+
"step": 24472
|
| 732 |
},
|
| 733 |
{
|
| 734 |
+
"epoch": 8.05,
|
| 735 |
+
"learning_rate": 5.151830685780705e-05,
|
| 736 |
+
"loss": 0.0002,
|
| 737 |
+
"step": 24610
|
| 738 |
},
|
| 739 |
{
|
| 740 |
+
"epoch": 8.12,
|
| 741 |
+
"learning_rate": 5.096135218907401e-05,
|
| 742 |
+
"loss": 0.0002,
|
| 743 |
+
"step": 24840
|
| 744 |
},
|
| 745 |
{
|
| 746 |
+
"epoch": 8.2,
|
| 747 |
+
"learning_rate": 5.040439752034095e-05,
|
| 748 |
+
"loss": 0.0002,
|
| 749 |
+
"step": 25070
|
| 750 |
},
|
| 751 |
{
|
| 752 |
+
"epoch": 8.27,
|
| 753 |
+
"learning_rate": 4.984744285160791e-05,
|
| 754 |
+
"loss": 0.0002,
|
| 755 |
+
"step": 25300
|
| 756 |
},
|
| 757 |
{
|
| 758 |
+
"epoch": 8.35,
|
| 759 |
+
"learning_rate": 4.929048818287485e-05,
|
| 760 |
+
"loss": 0.0002,
|
| 761 |
+
"step": 25530
|
|
|
|
|
|
|
|
|
|
|
|
|
| 762 |
},
|
| 763 |
{
|
| 764 |
+
"epoch": 8.42,
|
| 765 |
+
"learning_rate": 4.873353351414181e-05,
|
| 766 |
+
"loss": 0.0002,
|
| 767 |
+
"step": 25760
|
| 768 |
},
|
| 769 |
{
|
| 770 |
+
"epoch": 8.5,
|
| 771 |
+
"learning_rate": 4.817657884540876e-05,
|
| 772 |
+
"loss": 0.0002,
|
| 773 |
+
"step": 25990
|
| 774 |
},
|
| 775 |
{
|
| 776 |
+
"epoch": 8.57,
|
| 777 |
+
"learning_rate": 4.761962417667571e-05,
|
| 778 |
+
"loss": 0.0002,
|
| 779 |
+
"step": 26220
|
| 780 |
},
|
| 781 |
{
|
| 782 |
+
"epoch": 8.65,
|
| 783 |
+
"learning_rate": 4.706266950794266e-05,
|
| 784 |
+
"loss": 0.0002,
|
| 785 |
+
"step": 26450
|
| 786 |
},
|
| 787 |
{
|
| 788 |
+
"epoch": 8.72,
|
| 789 |
+
"learning_rate": 4.650571483920961e-05,
|
| 790 |
+
"loss": 0.0002,
|
| 791 |
+
"step": 26680
|
| 792 |
},
|
| 793 |
{
|
| 794 |
+
"epoch": 8.8,
|
| 795 |
+
"learning_rate": 4.594876017047656e-05,
|
| 796 |
+
"loss": 0.0002,
|
| 797 |
+
"step": 26910
|
| 798 |
},
|
| 799 |
{
|
| 800 |
+
"epoch": 8.87,
|
| 801 |
+
"learning_rate": 4.539180550174351e-05,
|
| 802 |
+
"loss": 0.0003,
|
| 803 |
+
"step": 27140
|
| 804 |
},
|
| 805 |
{
|
| 806 |
+
"epoch": 8.95,
|
| 807 |
+
"learning_rate": 4.483485083301046e-05,
|
| 808 |
+
"loss": 0.0002,
|
| 809 |
+
"step": 27370
|
| 810 |
},
|
| 811 |
{
|
| 812 |
+
"epoch": 9.0,
|
| 813 |
+
"eval_loss": 0.0007057118928059936,
|
| 814 |
"eval_max_distance": 3,
|
| 815 |
"eval_mean_distance": 0,
|
| 816 |
+
"eval_runtime": 30.795,
|
| 817 |
+
"eval_samples_per_second": 19.127,
|
| 818 |
+
"eval_steps_per_second": 0.13,
|
| 819 |
+
"step": 27531
|
| 820 |
},
|
| 821 |
{
|
| 822 |
+
"epoch": 9.02,
|
| 823 |
+
"learning_rate": 4.427789616427741e-05,
|
| 824 |
+
"loss": 0.0002,
|
| 825 |
+
"step": 27600
|
| 826 |
},
|
| 827 |
{
|
| 828 |
+
"epoch": 9.1,
|
| 829 |
+
"learning_rate": 4.372094149554436e-05,
|
| 830 |
+
"loss": 0.0002,
|
| 831 |
+
"step": 27830
|
| 832 |
},
|
| 833 |
{
|
| 834 |
+
"epoch": 9.17,
|
| 835 |
+
"learning_rate": 4.3163986826811313e-05,
|
| 836 |
+
"loss": 0.0002,
|
| 837 |
+
"step": 28060
|
| 838 |
},
|
| 839 |
{
|
| 840 |
+
"epoch": 9.25,
|
| 841 |
+
"learning_rate": 4.2607032158078264e-05,
|
| 842 |
+
"loss": 0.0002,
|
| 843 |
+
"step": 28290
|
| 844 |
},
|
| 845 |
{
|
| 846 |
+
"epoch": 9.32,
|
| 847 |
+
"learning_rate": 4.205007748934522e-05,
|
| 848 |
+
"loss": 0.0002,
|
| 849 |
+
"step": 28520
|
| 850 |
},
|
| 851 |
{
|
| 852 |
+
"epoch": 9.4,
|
| 853 |
+
"learning_rate": 4.149312282061217e-05,
|
| 854 |
+
"loss": 0.0002,
|
| 855 |
+
"step": 28750
|
| 856 |
},
|
| 857 |
{
|
| 858 |
+
"epoch": 9.47,
|
| 859 |
+
"learning_rate": 4.0936168151879115e-05,
|
| 860 |
+
"loss": 0.0002,
|
| 861 |
+
"step": 28980
|
| 862 |
},
|
| 863 |
{
|
| 864 |
+
"epoch": 9.55,
|
| 865 |
+
"learning_rate": 4.0379213483146065e-05,
|
| 866 |
+
"loss": 0.0002,
|
| 867 |
+
"step": 29210
|
|
|
|
|
|
|
|
|
|
|
|
|
| 868 |
},
|
| 869 |
{
|
| 870 |
+
"epoch": 9.62,
|
| 871 |
+
"learning_rate": 3.982225881441302e-05,
|
| 872 |
+
"loss": 0.0002,
|
| 873 |
+
"step": 29440
|
| 874 |
},
|
| 875 |
{
|
| 876 |
+
"epoch": 9.7,
|
| 877 |
+
"learning_rate": 3.926530414567997e-05,
|
| 878 |
"loss": 0.0003,
|
| 879 |
+
"step": 29670
|
| 880 |
},
|
| 881 |
{
|
| 882 |
+
"epoch": 9.77,
|
| 883 |
+
"learning_rate": 3.870834947694692e-05,
|
| 884 |
+
"loss": 0.0002,
|
| 885 |
+
"step": 29900
|
| 886 |
},
|
| 887 |
{
|
| 888 |
+
"epoch": 9.85,
|
| 889 |
+
"learning_rate": 3.8151394808213873e-05,
|
| 890 |
+
"loss": 0.0002,
|
| 891 |
+
"step": 30130
|
| 892 |
},
|
| 893 |
{
|
| 894 |
+
"epoch": 9.92,
|
| 895 |
+
"learning_rate": 3.7594440139480824e-05,
|
| 896 |
+
"loss": 0.0002,
|
| 897 |
+
"step": 30360
|
| 898 |
},
|
| 899 |
{
|
| 900 |
+
"epoch": 10.0,
|
| 901 |
+
"learning_rate": 3.7037485470747774e-05,
|
| 902 |
+
"loss": 0.0002,
|
| 903 |
+
"step": 30590
|
| 904 |
},
|
| 905 |
{
|
| 906 |
+
"epoch": 10.0,
|
| 907 |
+
"eval_loss": 0.000787499884609133,
|
| 908 |
+
"eval_max_distance": 3,
|
| 909 |
+
"eval_mean_distance": 0,
|
| 910 |
+
"eval_runtime": 30.7843,
|
| 911 |
+
"eval_samples_per_second": 19.133,
|
| 912 |
+
"eval_steps_per_second": 0.13,
|
| 913 |
+
"step": 30590
|
| 914 |
},
|
| 915 |
{
|
| 916 |
+
"epoch": 10.08,
|
| 917 |
+
"learning_rate": 3.6480530802014724e-05,
|
| 918 |
+
"loss": 0.0002,
|
| 919 |
+
"step": 30820
|
| 920 |
},
|
| 921 |
{
|
| 922 |
+
"epoch": 10.15,
|
| 923 |
+
"learning_rate": 3.5923576133281675e-05,
|
| 924 |
+
"loss": 0.0002,
|
| 925 |
+
"step": 31050
|
|
|
|
|
|
|
|
|
|
|
|
|
| 926 |
},
|
| 927 |
{
|
| 928 |
+
"epoch": 10.23,
|
| 929 |
+
"learning_rate": 3.5366621464548625e-05,
|
| 930 |
+
"loss": 0.0002,
|
| 931 |
+
"step": 31280
|
| 932 |
},
|
| 933 |
{
|
| 934 |
+
"epoch": 10.3,
|
| 935 |
+
"learning_rate": 3.4809666795815576e-05,
|
| 936 |
+
"loss": 0.0002,
|
| 937 |
+
"step": 31510
|
| 938 |
},
|
| 939 |
{
|
| 940 |
+
"epoch": 10.38,
|
| 941 |
+
"learning_rate": 3.4252712127082526e-05,
|
| 942 |
+
"loss": 0.0002,
|
| 943 |
+
"step": 31740
|
| 944 |
},
|
| 945 |
{
|
| 946 |
+
"epoch": 10.45,
|
| 947 |
+
"learning_rate": 3.369575745834948e-05,
|
| 948 |
+
"loss": 0.0002,
|
| 949 |
+
"step": 31970
|
| 950 |
},
|
| 951 |
{
|
| 952 |
+
"epoch": 10.53,
|
| 953 |
+
"learning_rate": 3.313880278961643e-05,
|
| 954 |
"loss": 0.0002,
|
| 955 |
+
"step": 32200
|
| 956 |
},
|
| 957 |
{
|
| 958 |
+
"epoch": 10.6,
|
| 959 |
+
"learning_rate": 3.258184812088338e-05,
|
| 960 |
+
"loss": 0.0002,
|
| 961 |
+
"step": 32430
|
| 962 |
},
|
| 963 |
{
|
| 964 |
+
"epoch": 10.68,
|
| 965 |
+
"learning_rate": 3.202489345215033e-05,
|
| 966 |
+
"loss": 0.0002,
|
| 967 |
+
"step": 32660
|
| 968 |
},
|
| 969 |
{
|
| 970 |
+
"epoch": 10.75,
|
| 971 |
+
"learning_rate": 3.1467938783417284e-05,
|
| 972 |
+
"loss": 0.0002,
|
| 973 |
+
"step": 32890
|
| 974 |
},
|
| 975 |
{
|
| 976 |
+
"epoch": 10.83,
|
| 977 |
+
"learning_rate": 3.0910984114684235e-05,
|
| 978 |
+
"loss": 0.0002,
|
| 979 |
+
"step": 33120
|
|
|
|
|
|
|
|
|
|
|
|
|
| 980 |
},
|
| 981 |
{
|
| 982 |
+
"epoch": 10.9,
|
| 983 |
+
"learning_rate": 3.0354029445951182e-05,
|
| 984 |
+
"loss": 0.0002,
|
| 985 |
+
"step": 33350
|
| 986 |
},
|
| 987 |
{
|
| 988 |
+
"epoch": 10.98,
|
| 989 |
+
"learning_rate": 2.9797074777218132e-05,
|
| 990 |
+
"loss": 0.0002,
|
| 991 |
+
"step": 33580
|
| 992 |
},
|
| 993 |
{
|
| 994 |
+
"epoch": 11.0,
|
| 995 |
+
"eval_loss": 0.000802784226834774,
|
| 996 |
+
"eval_max_distance": 3,
|
| 997 |
+
"eval_mean_distance": 0,
|
| 998 |
+
"eval_runtime": 30.8756,
|
| 999 |
+
"eval_samples_per_second": 19.077,
|
| 1000 |
+
"eval_steps_per_second": 0.13,
|
| 1001 |
+
"step": 33649
|
| 1002 |
},
|
| 1003 |
{
|
| 1004 |
+
"epoch": 11.05,
|
| 1005 |
+
"learning_rate": 2.9240120108485086e-05,
|
| 1006 |
+
"loss": 0.0002,
|
| 1007 |
+
"step": 33810
|
| 1008 |
},
|
| 1009 |
{
|
| 1010 |
+
"epoch": 11.13,
|
| 1011 |
+
"learning_rate": 2.8683165439752036e-05,
|
| 1012 |
+
"loss": 0.0002,
|
| 1013 |
+
"step": 34040
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
+
"epoch": 11.2,
|
| 1017 |
+
"learning_rate": 2.8126210771018983e-05,
|
| 1018 |
+
"loss": 0.0002,
|
| 1019 |
+
"step": 34270
|
| 1020 |
},
|
| 1021 |
{
|
| 1022 |
+
"epoch": 11.28,
|
| 1023 |
+
"learning_rate": 2.756925610228594e-05,
|
| 1024 |
+
"loss": 0.0002,
|
| 1025 |
+
"step": 34500
|
| 1026 |
},
|
| 1027 |
{
|
| 1028 |
+
"epoch": 11.35,
|
| 1029 |
+
"learning_rate": 2.7012301433552887e-05,
|
| 1030 |
+
"loss": 0.0002,
|
| 1031 |
+
"step": 34730
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1032 |
},
|
| 1033 |
{
|
| 1034 |
+
"epoch": 11.43,
|
| 1035 |
+
"learning_rate": 2.6455346764819838e-05,
|
| 1036 |
+
"loss": 0.0002,
|
| 1037 |
+
"step": 34960
|
| 1038 |
},
|
| 1039 |
{
|
| 1040 |
+
"epoch": 11.5,
|
| 1041 |
+
"learning_rate": 2.5898392096086788e-05,
|
| 1042 |
+
"loss": 0.0002,
|
| 1043 |
+
"step": 35190
|
| 1044 |
},
|
| 1045 |
{
|
| 1046 |
+
"epoch": 11.58,
|
| 1047 |
+
"learning_rate": 2.5341437427353742e-05,
|
| 1048 |
+
"loss": 0.0002,
|
| 1049 |
+
"step": 35420
|
| 1050 |
},
|
| 1051 |
{
|
| 1052 |
+
"epoch": 11.65,
|
| 1053 |
+
"learning_rate": 2.4784482758620692e-05,
|
| 1054 |
+
"loss": 0.0002,
|
| 1055 |
+
"step": 35650
|
| 1056 |
},
|
| 1057 |
{
|
| 1058 |
+
"epoch": 11.73,
|
| 1059 |
+
"learning_rate": 2.4227528089887643e-05,
|
| 1060 |
+
"loss": 0.0002,
|
| 1061 |
+
"step": 35880
|
| 1062 |
},
|
| 1063 |
{
|
| 1064 |
+
"epoch": 11.8,
|
| 1065 |
+
"learning_rate": 2.3670573421154593e-05,
|
| 1066 |
+
"loss": 0.0002,
|
| 1067 |
+
"step": 36110
|
| 1068 |
},
|
| 1069 |
{
|
| 1070 |
+
"epoch": 11.88,
|
| 1071 |
+
"learning_rate": 2.3113618752421543e-05,
|
| 1072 |
+
"loss": 0.0002,
|
| 1073 |
+
"step": 36340
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
+
"epoch": 11.95,
|
| 1077 |
+
"learning_rate": 2.2556664083688494e-05,
|
| 1078 |
+
"loss": 0.0002,
|
| 1079 |
+
"step": 36570
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
+
"epoch": 12.0,
|
| 1083 |
+
"eval_loss": 0.0007881763740442693,
|
| 1084 |
"eval_max_distance": 3,
|
| 1085 |
"eval_mean_distance": 0,
|
| 1086 |
+
"eval_runtime": 30.7194,
|
| 1087 |
+
"eval_samples_per_second": 19.174,
|
| 1088 |
+
"eval_steps_per_second": 0.13,
|
| 1089 |
+
"step": 36708
|
| 1090 |
},
|
| 1091 |
{
|
| 1092 |
+
"epoch": 12.03,
|
| 1093 |
+
"learning_rate": 2.1999709414955444e-05,
|
| 1094 |
+
"loss": 0.0002,
|
| 1095 |
+
"step": 36800
|
| 1096 |
},
|
| 1097 |
{
|
| 1098 |
+
"epoch": 12.11,
|
| 1099 |
+
"learning_rate": 2.1442754746222398e-05,
|
| 1100 |
+
"loss": 0.0002,
|
| 1101 |
+
"step": 37030
|
| 1102 |
},
|
| 1103 |
{
|
| 1104 |
+
"epoch": 12.18,
|
| 1105 |
+
"learning_rate": 2.0885800077489345e-05,
|
| 1106 |
+
"loss": 0.0002,
|
| 1107 |
+
"step": 37260
|
| 1108 |
},
|
| 1109 |
{
|
| 1110 |
+
"epoch": 12.26,
|
| 1111 |
+
"learning_rate": 2.03288454087563e-05,
|
| 1112 |
+
"loss": 0.0002,
|
| 1113 |
+
"step": 37490
|
| 1114 |
},
|
| 1115 |
{
|
| 1116 |
+
"epoch": 12.33,
|
| 1117 |
+
"learning_rate": 1.9771890740023245e-05,
|
| 1118 |
+
"loss": 0.0002,
|
| 1119 |
+
"step": 37720
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
+
"epoch": 12.41,
|
| 1123 |
+
"learning_rate": 1.92149360712902e-05,
|
| 1124 |
"loss": 0.0002,
|
| 1125 |
+
"step": 37950
|
| 1126 |
},
|
| 1127 |
{
|
| 1128 |
+
"epoch": 12.48,
|
| 1129 |
+
"learning_rate": 1.865798140255715e-05,
|
| 1130 |
+
"loss": 0.0002,
|
| 1131 |
+
"step": 38180
|
| 1132 |
},
|
| 1133 |
{
|
| 1134 |
+
"epoch": 12.56,
|
| 1135 |
+
"learning_rate": 1.81010267338241e-05,
|
| 1136 |
+
"loss": 0.0002,
|
| 1137 |
+
"step": 38410
|
| 1138 |
},
|
| 1139 |
{
|
| 1140 |
+
"epoch": 12.63,
|
| 1141 |
+
"learning_rate": 1.754407206509105e-05,
|
| 1142 |
+
"loss": 0.0002,
|
| 1143 |
+
"step": 38640
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
+
"epoch": 12.71,
|
| 1147 |
+
"learning_rate": 1.6987117396358e-05,
|
| 1148 |
+
"loss": 0.0002,
|
| 1149 |
+
"step": 38870
|
| 1150 |
},
|
| 1151 |
{
|
| 1152 |
+
"epoch": 12.78,
|
| 1153 |
+
"learning_rate": 1.6430162727624954e-05,
|
| 1154 |
+
"loss": 0.0002,
|
| 1155 |
+
"step": 39100
|
| 1156 |
},
|
| 1157 |
{
|
| 1158 |
+
"epoch": 12.86,
|
| 1159 |
+
"learning_rate": 1.5873208058891905e-05,
|
| 1160 |
+
"loss": 0.0002,
|
| 1161 |
+
"step": 39330
|
| 1162 |
},
|
| 1163 |
{
|
| 1164 |
+
"epoch": 12.93,
|
| 1165 |
+
"learning_rate": 1.5316253390158855e-05,
|
| 1166 |
+
"loss": 0.0002,
|
| 1167 |
+
"step": 39560
|
| 1168 |
},
|
| 1169 |
{
|
| 1170 |
+
"epoch": 13.0,
|
| 1171 |
+
"eval_loss": 0.000838827807456255,
|
| 1172 |
+
"eval_max_distance": 3,
|
| 1173 |
+
"eval_mean_distance": 0,
|
| 1174 |
+
"eval_runtime": 30.7487,
|
| 1175 |
+
"eval_samples_per_second": 19.155,
|
| 1176 |
+
"eval_steps_per_second": 0.13,
|
| 1177 |
+
"step": 39767
|
| 1178 |
},
|
| 1179 |
{
|
| 1180 |
+
"epoch": 13.01,
|
| 1181 |
+
"learning_rate": 1.4759298721425804e-05,
|
| 1182 |
+
"loss": 0.0002,
|
| 1183 |
+
"step": 39790
|
| 1184 |
},
|
| 1185 |
{
|
| 1186 |
+
"epoch": 13.08,
|
| 1187 |
+
"learning_rate": 1.4202344052692756e-05,
|
| 1188 |
+
"loss": 0.0002,
|
| 1189 |
+
"step": 40020
|
| 1190 |
},
|
| 1191 |
{
|
| 1192 |
+
"epoch": 13.16,
|
| 1193 |
+
"learning_rate": 1.3645389383959706e-05,
|
| 1194 |
+
"loss": 0.0002,
|
| 1195 |
+
"step": 40250
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1196 |
},
|
| 1197 |
{
|
| 1198 |
+
"epoch": 13.23,
|
| 1199 |
+
"learning_rate": 1.3088434715226658e-05,
|
| 1200 |
+
"loss": 0.0002,
|
| 1201 |
+
"step": 40480
|
| 1202 |
},
|
| 1203 |
{
|
| 1204 |
+
"epoch": 13.31,
|
| 1205 |
+
"learning_rate": 1.2531480046493607e-05,
|
| 1206 |
+
"loss": 0.0002,
|
| 1207 |
+
"step": 40710
|
| 1208 |
},
|
| 1209 |
{
|
| 1210 |
+
"epoch": 13.38,
|
| 1211 |
+
"learning_rate": 1.1974525377760557e-05,
|
| 1212 |
"loss": 0.0002,
|
| 1213 |
+
"step": 40940
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
+
"epoch": 13.46,
|
| 1217 |
+
"learning_rate": 1.141757070902751e-05,
|
| 1218 |
+
"loss": 0.0002,
|
| 1219 |
+
"step": 41170
|
| 1220 |
},
|
| 1221 |
{
|
| 1222 |
+
"epoch": 13.53,
|
| 1223 |
+
"learning_rate": 1.0860616040294461e-05,
|
| 1224 |
+
"loss": 0.0002,
|
| 1225 |
+
"step": 41400
|
| 1226 |
},
|
| 1227 |
{
|
| 1228 |
+
"epoch": 13.61,
|
| 1229 |
+
"learning_rate": 1.0303661371561412e-05,
|
| 1230 |
+
"loss": 0.0002,
|
| 1231 |
+
"step": 41630
|
| 1232 |
},
|
| 1233 |
{
|
| 1234 |
+
"epoch": 13.68,
|
| 1235 |
+
"learning_rate": 9.746706702828362e-06,
|
| 1236 |
+
"loss": 0.0002,
|
| 1237 |
+
"step": 41860
|
| 1238 |
},
|
| 1239 |
{
|
| 1240 |
+
"epoch": 13.76,
|
| 1241 |
+
"learning_rate": 9.189752034095312e-06,
|
| 1242 |
+
"loss": 0.0002,
|
| 1243 |
+
"step": 42090
|
| 1244 |
},
|
| 1245 |
{
|
| 1246 |
+
"epoch": 13.83,
|
| 1247 |
+
"learning_rate": 8.632797365362263e-06,
|
| 1248 |
+
"loss": 0.0002,
|
| 1249 |
+
"step": 42320
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1250 |
},
|
| 1251 |
{
|
| 1252 |
+
"epoch": 13.91,
|
| 1253 |
+
"learning_rate": 8.075842696629215e-06,
|
| 1254 |
+
"loss": 0.0002,
|
| 1255 |
+
"step": 42550
|
| 1256 |
},
|
| 1257 |
{
|
| 1258 |
+
"epoch": 13.98,
|
| 1259 |
+
"learning_rate": 7.518888027896165e-06,
|
| 1260 |
+
"loss": 0.0002,
|
| 1261 |
+
"step": 42780
|
| 1262 |
},
|
| 1263 |
{
|
| 1264 |
+
"epoch": 14.0,
|
| 1265 |
+
"eval_loss": 0.0008361997315660119,
|
| 1266 |
+
"eval_max_distance": 3,
|
| 1267 |
+
"eval_mean_distance": 0,
|
| 1268 |
+
"eval_runtime": 30.9876,
|
| 1269 |
+
"eval_samples_per_second": 19.008,
|
| 1270 |
+
"eval_steps_per_second": 0.129,
|
| 1271 |
+
"step": 42826
|
| 1272 |
},
|
| 1273 |
{
|
| 1274 |
+
"epoch": 14.06,
|
| 1275 |
+
"learning_rate": 6.9619333591631155e-06,
|
| 1276 |
+
"loss": 0.0002,
|
| 1277 |
+
"step": 43010
|
| 1278 |
},
|
| 1279 |
{
|
| 1280 |
+
"epoch": 14.14,
|
| 1281 |
+
"learning_rate": 6.404978690430067e-06,
|
| 1282 |
+
"loss": 0.0002,
|
| 1283 |
+
"step": 43240
|
| 1284 |
},
|
| 1285 |
{
|
| 1286 |
+
"epoch": 14.21,
|
| 1287 |
+
"learning_rate": 5.848024021697017e-06,
|
| 1288 |
+
"loss": 0.0002,
|
| 1289 |
+
"step": 43470
|
| 1290 |
},
|
| 1291 |
{
|
| 1292 |
+
"epoch": 14.29,
|
| 1293 |
+
"learning_rate": 5.291069352963967e-06,
|
| 1294 |
+
"loss": 0.0002,
|
| 1295 |
+
"step": 43700
|
| 1296 |
},
|
| 1297 |
{
|
| 1298 |
+
"epoch": 14.36,
|
| 1299 |
+
"learning_rate": 4.7341146842309186e-06,
|
| 1300 |
+
"loss": 0.0002,
|
| 1301 |
+
"step": 43930
|
| 1302 |
},
|
| 1303 |
{
|
| 1304 |
+
"epoch": 14.44,
|
| 1305 |
+
"learning_rate": 4.177160015497869e-06,
|
| 1306 |
+
"loss": 0.0002,
|
| 1307 |
+
"step": 44160
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1308 |
},
|
| 1309 |
{
|
| 1310 |
+
"epoch": 14.51,
|
| 1311 |
+
"learning_rate": 3.6202053467648197e-06,
|
| 1312 |
+
"loss": 0.0002,
|
| 1313 |
+
"step": 44390
|
| 1314 |
},
|
| 1315 |
{
|
| 1316 |
+
"epoch": 14.59,
|
| 1317 |
+
"learning_rate": 3.063250678031771e-06,
|
| 1318 |
+
"loss": 0.0002,
|
| 1319 |
+
"step": 44620
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
+
"epoch": 14.66,
|
| 1323 |
+
"learning_rate": 2.5062960092987217e-06,
|
| 1324 |
+
"loss": 0.0002,
|
| 1325 |
+
"step": 44850
|
| 1326 |
},
|
| 1327 |
{
|
| 1328 |
+
"epoch": 14.74,
|
| 1329 |
+
"learning_rate": 1.949341340565672e-06,
|
| 1330 |
+
"loss": 0.0002,
|
| 1331 |
+
"step": 45080
|
| 1332 |
},
|
| 1333 |
{
|
| 1334 |
+
"epoch": 14.81,
|
| 1335 |
+
"learning_rate": 1.392386671832623e-06,
|
| 1336 |
+
"loss": 0.0002,
|
| 1337 |
+
"step": 45310
|
| 1338 |
},
|
| 1339 |
{
|
| 1340 |
+
"epoch": 14.89,
|
| 1341 |
+
"learning_rate": 8.354320030995738e-07,
|
| 1342 |
+
"loss": 0.0002,
|
| 1343 |
+
"step": 45540
|
| 1344 |
},
|
| 1345 |
{
|
| 1346 |
+
"epoch": 14.96,
|
| 1347 |
+
"learning_rate": 2.784773343665246e-07,
|
| 1348 |
+
"loss": 0.0002,
|
| 1349 |
+
"step": 45770
|
| 1350 |
},
|
| 1351 |
{
|
| 1352 |
+
"epoch": 15.0,
|
| 1353 |
+
"eval_loss": 0.0008401814266107976,
|
| 1354 |
+
"eval_max_distance": 3,
|
| 1355 |
"eval_mean_distance": 0,
|
| 1356 |
+
"eval_runtime": 30.8673,
|
| 1357 |
+
"eval_samples_per_second": 19.082,
|
| 1358 |
+
"eval_steps_per_second": 0.13,
|
| 1359 |
+
"step": 45885
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 15.0,
|
| 1363 |
+
"step": 45885,
|
| 1364 |
+
"total_flos": 7.988818341631795e+16,
|
| 1365 |
+
"train_loss": 0.0002746593983470409,
|
| 1366 |
+
"train_runtime": 6476.6732,
|
| 1367 |
+
"train_samples_per_second": 531.266,
|
| 1368 |
+
"train_steps_per_second": 7.085
|
| 1369 |
}
|
| 1370 |
],
|
| 1371 |
+
"logging_steps": 230,
|
| 1372 |
+
"max_steps": 45885,
|
| 1373 |
+
"num_input_tokens_seen": 0,
|
| 1374 |
+
"num_train_epochs": 15,
|
| 1375 |
+
"save_steps": 4589,
|
| 1376 |
+
"total_flos": 7.988818341631795e+16,
|
| 1377 |
+
"train_batch_size": 75,
|
| 1378 |
"trial_name": null,
|
| 1379 |
"trial_params": null
|
| 1380 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1eb73bbd8d63a4a4a349732132a9ced8e0528dece7669b9d0a3ba1f248d5ec37
|
| 3 |
+
size 4792
|