Upload 9 files
Browse files- config.json +28 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- rng_state (1).pth +3 -0
- scheduler (1).pt +3 -0
- special_tokens_map (1).json +15 -0
- tokenizer (1).json +0 -0
- tokenizer_config (1).json +15 -0
- trainer_state (1).json +647 -0
config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "roberta-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "roberta",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"problem_type": "single_label_classification",
|
| 23 |
+
"torch_dtype": "float32",
|
| 24 |
+
"transformers_version": "4.32.0",
|
| 25 |
+
"type_vocab_size": 1,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 50265
|
| 28 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17818305304f7d4deec1dad6f75442a6f38d2ee207be4ec007921a4e0b503bc2
|
| 3 |
+
size 498657649
|
rng_state (1).pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:69d4c9e61204c9303bed7a9b5edea1800fdeb30a0b194e441931c3b0d21eb9e6
|
| 3 |
+
size 14575
|
scheduler (1).pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:73c4fb322646cfd99e35d6c6e8c20b3eab6fdafe3db029c486caabde4c4589ab
|
| 3 |
+
size 627
|
special_tokens_map (1).json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
tokenizer (1).json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config (1).json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"bos_token": "<s>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"cls_token": "<s>",
|
| 6 |
+
"eos_token": "</s>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"mask_token": "<mask>",
|
| 9 |
+
"model_max_length": 512,
|
| 10 |
+
"pad_token": "<pad>",
|
| 11 |
+
"sep_token": "</s>",
|
| 12 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 13 |
+
"trim_offsets": true,
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
trainer_state (1).json
ADDED
|
@@ -0,0 +1,647 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.14040808379650116,
|
| 3 |
+
"best_model_checkpoint": "./results/checkpoint-51064",
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 51064,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"learning_rate": 1.986944487962818e-05,
|
| 14 |
+
"loss": 0.2496,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.04,
|
| 19 |
+
"learning_rate": 1.9738889759256358e-05,
|
| 20 |
+
"loss": 0.208,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.06,
|
| 25 |
+
"learning_rate": 1.9608334638884537e-05,
|
| 26 |
+
"loss": 0.1884,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.08,
|
| 31 |
+
"learning_rate": 1.9477779518512717e-05,
|
| 32 |
+
"loss": 0.1955,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.1,
|
| 37 |
+
"learning_rate": 1.9347224398140897e-05,
|
| 38 |
+
"loss": 0.2005,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.12,
|
| 43 |
+
"learning_rate": 1.9216669277769077e-05,
|
| 44 |
+
"loss": 0.1625,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.14,
|
| 49 |
+
"learning_rate": 1.9086114157397256e-05,
|
| 50 |
+
"loss": 0.18,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.16,
|
| 55 |
+
"learning_rate": 1.8955559037025433e-05,
|
| 56 |
+
"loss": 0.153,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.18,
|
| 61 |
+
"learning_rate": 1.8825003916653613e-05,
|
| 62 |
+
"loss": 0.1996,
|
| 63 |
+
"step": 4500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.2,
|
| 67 |
+
"learning_rate": 1.8694448796281792e-05,
|
| 68 |
+
"loss": 0.1643,
|
| 69 |
+
"step": 5000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.22,
|
| 73 |
+
"learning_rate": 1.856389367590997e-05,
|
| 74 |
+
"loss": 0.2022,
|
| 75 |
+
"step": 5500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.23,
|
| 79 |
+
"learning_rate": 1.843333855553815e-05,
|
| 80 |
+
"loss": 0.1706,
|
| 81 |
+
"step": 6000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.25,
|
| 85 |
+
"learning_rate": 1.8302783435166328e-05,
|
| 86 |
+
"loss": 0.1895,
|
| 87 |
+
"step": 6500
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.27,
|
| 91 |
+
"learning_rate": 1.8172228314794508e-05,
|
| 92 |
+
"loss": 0.1733,
|
| 93 |
+
"step": 7000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.29,
|
| 97 |
+
"learning_rate": 1.8041673194422688e-05,
|
| 98 |
+
"loss": 0.1938,
|
| 99 |
+
"step": 7500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.31,
|
| 103 |
+
"learning_rate": 1.7911118074050867e-05,
|
| 104 |
+
"loss": 0.1786,
|
| 105 |
+
"step": 8000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.33,
|
| 109 |
+
"learning_rate": 1.7780562953679047e-05,
|
| 110 |
+
"loss": 0.19,
|
| 111 |
+
"step": 8500
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.35,
|
| 115 |
+
"learning_rate": 1.7650007833307223e-05,
|
| 116 |
+
"loss": 0.1908,
|
| 117 |
+
"step": 9000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.37,
|
| 121 |
+
"learning_rate": 1.7519452712935403e-05,
|
| 122 |
+
"loss": 0.1755,
|
| 123 |
+
"step": 9500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.39,
|
| 127 |
+
"learning_rate": 1.7388897592563583e-05,
|
| 128 |
+
"loss": 0.1536,
|
| 129 |
+
"step": 10000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.41,
|
| 133 |
+
"learning_rate": 1.725834247219176e-05,
|
| 134 |
+
"loss": 0.1574,
|
| 135 |
+
"step": 10500
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.43,
|
| 139 |
+
"learning_rate": 1.712778735181994e-05,
|
| 140 |
+
"loss": 0.1472,
|
| 141 |
+
"step": 11000
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.45,
|
| 145 |
+
"learning_rate": 1.699723223144812e-05,
|
| 146 |
+
"loss": 0.1936,
|
| 147 |
+
"step": 11500
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.47,
|
| 151 |
+
"learning_rate": 1.68666771110763e-05,
|
| 152 |
+
"loss": 0.1746,
|
| 153 |
+
"step": 12000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.49,
|
| 157 |
+
"learning_rate": 1.6736121990704478e-05,
|
| 158 |
+
"loss": 0.1731,
|
| 159 |
+
"step": 12500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.51,
|
| 163 |
+
"learning_rate": 1.6605566870332658e-05,
|
| 164 |
+
"loss": 0.1927,
|
| 165 |
+
"step": 13000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.53,
|
| 169 |
+
"learning_rate": 1.6475011749960834e-05,
|
| 170 |
+
"loss": 0.1946,
|
| 171 |
+
"step": 13500
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.55,
|
| 175 |
+
"learning_rate": 1.6344456629589014e-05,
|
| 176 |
+
"loss": 0.1636,
|
| 177 |
+
"step": 14000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.57,
|
| 181 |
+
"learning_rate": 1.6213901509217194e-05,
|
| 182 |
+
"loss": 0.2385,
|
| 183 |
+
"step": 14500
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.59,
|
| 187 |
+
"learning_rate": 1.608334638884537e-05,
|
| 188 |
+
"loss": 0.2137,
|
| 189 |
+
"step": 15000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.61,
|
| 193 |
+
"learning_rate": 1.595279126847355e-05,
|
| 194 |
+
"loss": 0.1736,
|
| 195 |
+
"step": 15500
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.63,
|
| 199 |
+
"learning_rate": 1.582223614810173e-05,
|
| 200 |
+
"loss": 0.2196,
|
| 201 |
+
"step": 16000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.65,
|
| 205 |
+
"learning_rate": 1.569168102772991e-05,
|
| 206 |
+
"loss": 0.1878,
|
| 207 |
+
"step": 16500
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.67,
|
| 211 |
+
"learning_rate": 1.556112590735809e-05,
|
| 212 |
+
"loss": 0.1742,
|
| 213 |
+
"step": 17000
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.69,
|
| 217 |
+
"learning_rate": 1.543057078698627e-05,
|
| 218 |
+
"loss": 0.1907,
|
| 219 |
+
"step": 17500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.7,
|
| 223 |
+
"learning_rate": 1.5300015666614445e-05,
|
| 224 |
+
"loss": 0.2048,
|
| 225 |
+
"step": 18000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.72,
|
| 229 |
+
"learning_rate": 1.5169460546242625e-05,
|
| 230 |
+
"loss": 0.1777,
|
| 231 |
+
"step": 18500
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.74,
|
| 235 |
+
"learning_rate": 1.5038905425870805e-05,
|
| 236 |
+
"loss": 0.213,
|
| 237 |
+
"step": 19000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.76,
|
| 241 |
+
"learning_rate": 1.4908350305498983e-05,
|
| 242 |
+
"loss": 0.1941,
|
| 243 |
+
"step": 19500
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.78,
|
| 247 |
+
"learning_rate": 1.4777795185127163e-05,
|
| 248 |
+
"loss": 0.1525,
|
| 249 |
+
"step": 20000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.8,
|
| 253 |
+
"learning_rate": 1.4647240064755342e-05,
|
| 254 |
+
"loss": 0.1638,
|
| 255 |
+
"step": 20500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.82,
|
| 259 |
+
"learning_rate": 1.4516684944383519e-05,
|
| 260 |
+
"loss": 0.174,
|
| 261 |
+
"step": 21000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.84,
|
| 265 |
+
"learning_rate": 1.4386129824011698e-05,
|
| 266 |
+
"loss": 0.1693,
|
| 267 |
+
"step": 21500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.86,
|
| 271 |
+
"learning_rate": 1.4255574703639878e-05,
|
| 272 |
+
"loss": 0.1926,
|
| 273 |
+
"step": 22000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.88,
|
| 277 |
+
"learning_rate": 1.4125019583268056e-05,
|
| 278 |
+
"loss": 0.1915,
|
| 279 |
+
"step": 22500
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.9,
|
| 283 |
+
"learning_rate": 1.3994464462896236e-05,
|
| 284 |
+
"loss": 0.1671,
|
| 285 |
+
"step": 23000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.92,
|
| 289 |
+
"learning_rate": 1.3863909342524416e-05,
|
| 290 |
+
"loss": 0.1572,
|
| 291 |
+
"step": 23500
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 0.94,
|
| 295 |
+
"learning_rate": 1.3733354222152594e-05,
|
| 296 |
+
"loss": 0.1613,
|
| 297 |
+
"step": 24000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.96,
|
| 301 |
+
"learning_rate": 1.3602799101780773e-05,
|
| 302 |
+
"loss": 0.165,
|
| 303 |
+
"step": 24500
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.98,
|
| 307 |
+
"learning_rate": 1.3472243981408951e-05,
|
| 308 |
+
"loss": 0.1739,
|
| 309 |
+
"step": 25000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.0,
|
| 313 |
+
"learning_rate": 1.3341688861037131e-05,
|
| 314 |
+
"loss": 0.1611,
|
| 315 |
+
"step": 25500
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 1.0,
|
| 319 |
+
"eval_loss": 0.17089270055294037,
|
| 320 |
+
"eval_runtime": 899.1592,
|
| 321 |
+
"eval_samples_per_second": 35.494,
|
| 322 |
+
"eval_steps_per_second": 7.099,
|
| 323 |
+
"step": 25532
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 1.02,
|
| 327 |
+
"learning_rate": 1.321113374066531e-05,
|
| 328 |
+
"loss": 0.1776,
|
| 329 |
+
"step": 26000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 1.04,
|
| 333 |
+
"learning_rate": 1.3080578620293489e-05,
|
| 334 |
+
"loss": 0.1759,
|
| 335 |
+
"step": 26500
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 1.06,
|
| 339 |
+
"learning_rate": 1.2950023499921669e-05,
|
| 340 |
+
"loss": 0.1544,
|
| 341 |
+
"step": 27000
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 1.08,
|
| 345 |
+
"learning_rate": 1.2819468379549847e-05,
|
| 346 |
+
"loss": 0.1691,
|
| 347 |
+
"step": 27500
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 1.1,
|
| 351 |
+
"learning_rate": 1.2688913259178027e-05,
|
| 352 |
+
"loss": 0.1866,
|
| 353 |
+
"step": 28000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.12,
|
| 357 |
+
"learning_rate": 1.2558358138806206e-05,
|
| 358 |
+
"loss": 0.1833,
|
| 359 |
+
"step": 28500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.14,
|
| 363 |
+
"learning_rate": 1.2427803018434383e-05,
|
| 364 |
+
"loss": 0.1791,
|
| 365 |
+
"step": 29000
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.16,
|
| 369 |
+
"learning_rate": 1.2297247898062562e-05,
|
| 370 |
+
"loss": 0.173,
|
| 371 |
+
"step": 29500
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 1.17,
|
| 375 |
+
"learning_rate": 1.2166692777690742e-05,
|
| 376 |
+
"loss": 0.2536,
|
| 377 |
+
"step": 30000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.19,
|
| 381 |
+
"learning_rate": 1.203613765731892e-05,
|
| 382 |
+
"loss": 0.2705,
|
| 383 |
+
"step": 30500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.21,
|
| 387 |
+
"learning_rate": 1.19055825369471e-05,
|
| 388 |
+
"loss": 0.2602,
|
| 389 |
+
"step": 31000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.23,
|
| 393 |
+
"learning_rate": 1.177502741657528e-05,
|
| 394 |
+
"loss": 0.1628,
|
| 395 |
+
"step": 31500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.25,
|
| 399 |
+
"learning_rate": 1.1644472296203458e-05,
|
| 400 |
+
"loss": 0.249,
|
| 401 |
+
"step": 32000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.27,
|
| 405 |
+
"learning_rate": 1.1513917175831637e-05,
|
| 406 |
+
"loss": 0.2991,
|
| 407 |
+
"step": 32500
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 1.29,
|
| 411 |
+
"learning_rate": 1.1383362055459817e-05,
|
| 412 |
+
"loss": 0.3618,
|
| 413 |
+
"step": 33000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.31,
|
| 417 |
+
"learning_rate": 1.1252806935087994e-05,
|
| 418 |
+
"loss": 0.3487,
|
| 419 |
+
"step": 33500
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 1.33,
|
| 423 |
+
"learning_rate": 1.1122251814716173e-05,
|
| 424 |
+
"loss": 0.3179,
|
| 425 |
+
"step": 34000
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 1.35,
|
| 429 |
+
"learning_rate": 1.0991696694344353e-05,
|
| 430 |
+
"loss": 0.3457,
|
| 431 |
+
"step": 34500
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 1.37,
|
| 435 |
+
"learning_rate": 1.0861141573972531e-05,
|
| 436 |
+
"loss": 0.3574,
|
| 437 |
+
"step": 35000
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.39,
|
| 441 |
+
"learning_rate": 1.073058645360071e-05,
|
| 442 |
+
"loss": 0.3486,
|
| 443 |
+
"step": 35500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.41,
|
| 447 |
+
"learning_rate": 1.060003133322889e-05,
|
| 448 |
+
"loss": 0.2576,
|
| 449 |
+
"step": 36000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 1.43,
|
| 453 |
+
"learning_rate": 1.0469476212857069e-05,
|
| 454 |
+
"loss": 0.1753,
|
| 455 |
+
"step": 36500
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 1.45,
|
| 459 |
+
"learning_rate": 1.0338921092485248e-05,
|
| 460 |
+
"loss": 0.2012,
|
| 461 |
+
"step": 37000
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 1.47,
|
| 465 |
+
"learning_rate": 1.0208365972113428e-05,
|
| 466 |
+
"loss": 0.2095,
|
| 467 |
+
"step": 37500
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 1.49,
|
| 471 |
+
"learning_rate": 1.0077810851741604e-05,
|
| 472 |
+
"loss": 0.1966,
|
| 473 |
+
"step": 38000
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.51,
|
| 477 |
+
"learning_rate": 9.947255731369784e-06,
|
| 478 |
+
"loss": 0.1492,
|
| 479 |
+
"step": 38500
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.53,
|
| 483 |
+
"learning_rate": 9.816700610997964e-06,
|
| 484 |
+
"loss": 0.1663,
|
| 485 |
+
"step": 39000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.55,
|
| 489 |
+
"learning_rate": 9.686145490626144e-06,
|
| 490 |
+
"loss": 0.158,
|
| 491 |
+
"step": 39500
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 1.57,
|
| 495 |
+
"learning_rate": 9.555590370254322e-06,
|
| 496 |
+
"loss": 0.1635,
|
| 497 |
+
"step": 40000
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.59,
|
| 501 |
+
"learning_rate": 9.425035249882501e-06,
|
| 502 |
+
"loss": 0.1605,
|
| 503 |
+
"step": 40500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 1.61,
|
| 507 |
+
"learning_rate": 9.29448012951068e-06,
|
| 508 |
+
"loss": 0.1479,
|
| 509 |
+
"step": 41000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 1.63,
|
| 513 |
+
"learning_rate": 9.16392500913886e-06,
|
| 514 |
+
"loss": 0.1555,
|
| 515 |
+
"step": 41500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 1.64,
|
| 519 |
+
"learning_rate": 9.033369888767039e-06,
|
| 520 |
+
"loss": 0.1657,
|
| 521 |
+
"step": 42000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.66,
|
| 525 |
+
"learning_rate": 8.902814768395217e-06,
|
| 526 |
+
"loss": 0.1322,
|
| 527 |
+
"step": 42500
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.68,
|
| 531 |
+
"learning_rate": 8.772259648023395e-06,
|
| 532 |
+
"loss": 0.1441,
|
| 533 |
+
"step": 43000
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 1.7,
|
| 537 |
+
"learning_rate": 8.641704527651575e-06,
|
| 538 |
+
"loss": 0.1514,
|
| 539 |
+
"step": 43500
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 1.72,
|
| 543 |
+
"learning_rate": 8.511149407279755e-06,
|
| 544 |
+
"loss": 0.1556,
|
| 545 |
+
"step": 44000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 1.74,
|
| 549 |
+
"learning_rate": 8.380594286907934e-06,
|
| 550 |
+
"loss": 0.2028,
|
| 551 |
+
"step": 44500
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 1.76,
|
| 555 |
+
"learning_rate": 8.250039166536112e-06,
|
| 556 |
+
"loss": 0.1538,
|
| 557 |
+
"step": 45000
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.78,
|
| 561 |
+
"learning_rate": 8.11948404616429e-06,
|
| 562 |
+
"loss": 0.1804,
|
| 563 |
+
"step": 45500
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.8,
|
| 567 |
+
"learning_rate": 7.98892892579247e-06,
|
| 568 |
+
"loss": 0.146,
|
| 569 |
+
"step": 46000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.82,
|
| 573 |
+
"learning_rate": 7.85837380542065e-06,
|
| 574 |
+
"loss": 0.1538,
|
| 575 |
+
"step": 46500
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 1.84,
|
| 579 |
+
"learning_rate": 7.727818685048828e-06,
|
| 580 |
+
"loss": 0.1476,
|
| 581 |
+
"step": 47000
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 1.86,
|
| 585 |
+
"learning_rate": 7.597263564677008e-06,
|
| 586 |
+
"loss": 0.1766,
|
| 587 |
+
"step": 47500
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 1.88,
|
| 591 |
+
"learning_rate": 7.4667084443051865e-06,
|
| 592 |
+
"loss": 0.1807,
|
| 593 |
+
"step": 48000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 1.9,
|
| 597 |
+
"learning_rate": 7.336153323933365e-06,
|
| 598 |
+
"loss": 0.1416,
|
| 599 |
+
"step": 48500
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 1.92,
|
| 603 |
+
"learning_rate": 7.205598203561544e-06,
|
| 604 |
+
"loss": 0.1453,
|
| 605 |
+
"step": 49000
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.94,
|
| 609 |
+
"learning_rate": 7.075043083189723e-06,
|
| 610 |
+
"loss": 0.1349,
|
| 611 |
+
"step": 49500
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.96,
|
| 615 |
+
"learning_rate": 6.944487962817902e-06,
|
| 616 |
+
"loss": 0.1852,
|
| 617 |
+
"step": 50000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 1.98,
|
| 621 |
+
"learning_rate": 6.813932842446082e-06,
|
| 622 |
+
"loss": 0.1599,
|
| 623 |
+
"step": 50500
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 2.0,
|
| 627 |
+
"learning_rate": 6.68337772207426e-06,
|
| 628 |
+
"loss": 0.1184,
|
| 629 |
+
"step": 51000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 2.0,
|
| 633 |
+
"eval_loss": 0.14040808379650116,
|
| 634 |
+
"eval_runtime": 896.823,
|
| 635 |
+
"eval_samples_per_second": 35.587,
|
| 636 |
+
"eval_steps_per_second": 7.117,
|
| 637 |
+
"step": 51064
|
| 638 |
+
}
|
| 639 |
+
],
|
| 640 |
+
"logging_steps": 500,
|
| 641 |
+
"max_steps": 76596,
|
| 642 |
+
"num_train_epochs": 3,
|
| 643 |
+
"save_steps": 500,
|
| 644 |
+
"total_flos": 6.717540976607232e+16,
|
| 645 |
+
"trial_name": null,
|
| 646 |
+
"trial_params": null
|
| 647 |
+
}
|