Upload folder using huggingface_hub
Browse files- config.json +27 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +15 -0
- tokenizer.json +0 -0
- tokenizer_config.json +15 -0
- trainer_state.json +568 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/mnt/data/yule/.cache/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 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.33.1",
|
| 24 |
+
"type_vocab_size": 1,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 50265
|
| 27 |
+
}
|
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 sha256:5dfae9d6ad8ec419a2c2a24f3ba6ddcdbec744632baab69887048de64d4b137a
|
| 3 |
+
size 498654833
|
special_tokens_map.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.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.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": 1000000000000000019884624838656,
|
| 10 |
+
"pad_token": "<pad>",
|
| 11 |
+
"sep_token": "</s>",
|
| 12 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 13 |
+
"trim_offsets": true,
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,568 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9530809205642168,
|
| 3 |
+
"best_model_checkpoint": "./save_models/sst2/roberta-base_lr1e-05_run1/checkpoint-22734",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 37890,
|
| 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,
|
| 13 |
+
"learning_rate": 2.198768689533861e-06,
|
| 14 |
+
"loss": 0.6593,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.26,
|
| 19 |
+
"learning_rate": 4.397537379067722e-06,
|
| 20 |
+
"loss": 0.3243,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.4,
|
| 25 |
+
"learning_rate": 6.596306068601583e-06,
|
| 26 |
+
"loss": 0.2873,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.53,
|
| 31 |
+
"learning_rate": 8.795074758135444e-06,
|
| 32 |
+
"loss": 0.2614,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.66,
|
| 37 |
+
"learning_rate": 9.936545372866128e-06,
|
| 38 |
+
"loss": 0.2533,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.79,
|
| 43 |
+
"learning_rate": 9.796159029649596e-06,
|
| 44 |
+
"loss": 0.2394,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.92,
|
| 49 |
+
"learning_rate": 9.655772686433064e-06,
|
| 50 |
+
"loss": 0.2448,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 1.0,
|
| 55 |
+
"eval_accuracy": 0.9278396436525612,
|
| 56 |
+
"eval_loss": 0.25612303614616394,
|
| 57 |
+
"eval_runtime": 4.9386,
|
| 58 |
+
"eval_samples_per_second": 1363.746,
|
| 59 |
+
"eval_steps_per_second": 85.247,
|
| 60 |
+
"step": 3789
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.06,
|
| 64 |
+
"learning_rate": 9.515386343216533e-06,
|
| 65 |
+
"loss": 0.2235,
|
| 66 |
+
"step": 4000
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.19,
|
| 70 |
+
"learning_rate": 9.375000000000001e-06,
|
| 71 |
+
"loss": 0.1893,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.32,
|
| 76 |
+
"learning_rate": 9.234613656783469e-06,
|
| 77 |
+
"loss": 0.2059,
|
| 78 |
+
"step": 5000
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.45,
|
| 82 |
+
"learning_rate": 9.094227313566937e-06,
|
| 83 |
+
"loss": 0.2096,
|
| 84 |
+
"step": 5500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.58,
|
| 88 |
+
"learning_rate": 8.953840970350406e-06,
|
| 89 |
+
"loss": 0.1936,
|
| 90 |
+
"step": 6000
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.72,
|
| 94 |
+
"learning_rate": 8.813454627133872e-06,
|
| 95 |
+
"loss": 0.2067,
|
| 96 |
+
"step": 6500
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.85,
|
| 100 |
+
"learning_rate": 8.67306828391734e-06,
|
| 101 |
+
"loss": 0.1879,
|
| 102 |
+
"step": 7000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.98,
|
| 106 |
+
"learning_rate": 8.532681940700809e-06,
|
| 107 |
+
"loss": 0.2037,
|
| 108 |
+
"step": 7500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 2.0,
|
| 112 |
+
"eval_accuracy": 0.948181143281366,
|
| 113 |
+
"eval_loss": 0.17774701118469238,
|
| 114 |
+
"eval_runtime": 4.9236,
|
| 115 |
+
"eval_samples_per_second": 1367.913,
|
| 116 |
+
"eval_steps_per_second": 85.507,
|
| 117 |
+
"step": 7578
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 2.11,
|
| 121 |
+
"learning_rate": 8.392295597484278e-06,
|
| 122 |
+
"loss": 0.1544,
|
| 123 |
+
"step": 8000
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 2.24,
|
| 127 |
+
"learning_rate": 8.251909254267747e-06,
|
| 128 |
+
"loss": 0.1502,
|
| 129 |
+
"step": 8500
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 2.38,
|
| 133 |
+
"learning_rate": 8.111522911051213e-06,
|
| 134 |
+
"loss": 0.1551,
|
| 135 |
+
"step": 9000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 2.51,
|
| 139 |
+
"learning_rate": 7.971136567834682e-06,
|
| 140 |
+
"loss": 0.1464,
|
| 141 |
+
"step": 9500
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 2.64,
|
| 145 |
+
"learning_rate": 7.83075022461815e-06,
|
| 146 |
+
"loss": 0.1668,
|
| 147 |
+
"step": 10000
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.77,
|
| 151 |
+
"learning_rate": 7.690363881401618e-06,
|
| 152 |
+
"loss": 0.1452,
|
| 153 |
+
"step": 10500
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.9,
|
| 157 |
+
"learning_rate": 7.549977538185086e-06,
|
| 158 |
+
"loss": 0.1558,
|
| 159 |
+
"step": 11000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 3.0,
|
| 163 |
+
"eval_accuracy": 0.9496659242761692,
|
| 164 |
+
"eval_loss": 0.2051115483045578,
|
| 165 |
+
"eval_runtime": 4.9481,
|
| 166 |
+
"eval_samples_per_second": 1361.12,
|
| 167 |
+
"eval_steps_per_second": 85.083,
|
| 168 |
+
"step": 11367
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 3.04,
|
| 172 |
+
"learning_rate": 7.409591194968554e-06,
|
| 173 |
+
"loss": 0.1437,
|
| 174 |
+
"step": 11500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 3.17,
|
| 178 |
+
"learning_rate": 7.269204851752022e-06,
|
| 179 |
+
"loss": 0.1174,
|
| 180 |
+
"step": 12000
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 3.3,
|
| 184 |
+
"learning_rate": 7.12881850853549e-06,
|
| 185 |
+
"loss": 0.135,
|
| 186 |
+
"step": 12500
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 3.43,
|
| 190 |
+
"learning_rate": 6.988432165318958e-06,
|
| 191 |
+
"loss": 0.126,
|
| 192 |
+
"step": 13000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 3.56,
|
| 196 |
+
"learning_rate": 6.848045822102426e-06,
|
| 197 |
+
"loss": 0.1249,
|
| 198 |
+
"step": 13500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 3.69,
|
| 202 |
+
"learning_rate": 6.707659478885895e-06,
|
| 203 |
+
"loss": 0.1366,
|
| 204 |
+
"step": 14000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 3.83,
|
| 208 |
+
"learning_rate": 6.567273135669363e-06,
|
| 209 |
+
"loss": 0.1355,
|
| 210 |
+
"step": 14500
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 3.96,
|
| 214 |
+
"learning_rate": 6.426886792452831e-06,
|
| 215 |
+
"loss": 0.1239,
|
| 216 |
+
"step": 15000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 4.0,
|
| 220 |
+
"eval_accuracy": 0.9512991833704528,
|
| 221 |
+
"eval_loss": 0.2515639364719391,
|
| 222 |
+
"eval_runtime": 4.9353,
|
| 223 |
+
"eval_samples_per_second": 1364.649,
|
| 224 |
+
"eval_steps_per_second": 85.303,
|
| 225 |
+
"step": 15156
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 4.09,
|
| 229 |
+
"learning_rate": 6.2865004492362994e-06,
|
| 230 |
+
"loss": 0.0957,
|
| 231 |
+
"step": 15500
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 4.22,
|
| 235 |
+
"learning_rate": 6.146114106019767e-06,
|
| 236 |
+
"loss": 0.1014,
|
| 237 |
+
"step": 16000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 4.35,
|
| 241 |
+
"learning_rate": 6.005727762803235e-06,
|
| 242 |
+
"loss": 0.0923,
|
| 243 |
+
"step": 16500
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 4.49,
|
| 247 |
+
"learning_rate": 5.865341419586703e-06,
|
| 248 |
+
"loss": 0.1169,
|
| 249 |
+
"step": 17000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 4.62,
|
| 253 |
+
"learning_rate": 5.7249550763701715e-06,
|
| 254 |
+
"loss": 0.1123,
|
| 255 |
+
"step": 17500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 4.75,
|
| 259 |
+
"learning_rate": 5.584568733153639e-06,
|
| 260 |
+
"loss": 0.1148,
|
| 261 |
+
"step": 18000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 4.88,
|
| 265 |
+
"learning_rate": 5.444182389937107e-06,
|
| 266 |
+
"loss": 0.109,
|
| 267 |
+
"step": 18500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 5.0,
|
| 271 |
+
"eval_accuracy": 0.9508537490720119,
|
| 272 |
+
"eval_loss": 0.2747756838798523,
|
| 273 |
+
"eval_runtime": 4.9452,
|
| 274 |
+
"eval_samples_per_second": 1361.92,
|
| 275 |
+
"eval_steps_per_second": 85.133,
|
| 276 |
+
"step": 18945
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 5.01,
|
| 280 |
+
"learning_rate": 5.303796046720575e-06,
|
| 281 |
+
"loss": 0.0971,
|
| 282 |
+
"step": 19000
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 5.15,
|
| 286 |
+
"learning_rate": 5.163409703504043e-06,
|
| 287 |
+
"loss": 0.0867,
|
| 288 |
+
"step": 19500
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 5.28,
|
| 292 |
+
"learning_rate": 5.023023360287511e-06,
|
| 293 |
+
"loss": 0.0813,
|
| 294 |
+
"step": 20000
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 5.41,
|
| 298 |
+
"learning_rate": 4.88263701707098e-06,
|
| 299 |
+
"loss": 0.0938,
|
| 300 |
+
"step": 20500
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 5.54,
|
| 304 |
+
"learning_rate": 4.7422506738544475e-06,
|
| 305 |
+
"loss": 0.0912,
|
| 306 |
+
"step": 21000
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 5.67,
|
| 310 |
+
"learning_rate": 4.601864330637916e-06,
|
| 311 |
+
"loss": 0.0787,
|
| 312 |
+
"step": 21500
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 5.81,
|
| 316 |
+
"learning_rate": 4.461477987421384e-06,
|
| 317 |
+
"loss": 0.0773,
|
| 318 |
+
"step": 22000
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 5.94,
|
| 322 |
+
"learning_rate": 4.321091644204852e-06,
|
| 323 |
+
"loss": 0.0915,
|
| 324 |
+
"step": 22500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 6.0,
|
| 328 |
+
"eval_accuracy": 0.9530809205642168,
|
| 329 |
+
"eval_loss": 0.27420464158058167,
|
| 330 |
+
"eval_runtime": 4.9297,
|
| 331 |
+
"eval_samples_per_second": 1366.207,
|
| 332 |
+
"eval_steps_per_second": 85.401,
|
| 333 |
+
"step": 22734
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 6.07,
|
| 337 |
+
"learning_rate": 4.18070530098832e-06,
|
| 338 |
+
"loss": 0.0761,
|
| 339 |
+
"step": 23000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 6.2,
|
| 343 |
+
"learning_rate": 4.040318957771789e-06,
|
| 344 |
+
"loss": 0.0694,
|
| 345 |
+
"step": 23500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 6.33,
|
| 349 |
+
"learning_rate": 3.899932614555256e-06,
|
| 350 |
+
"loss": 0.0657,
|
| 351 |
+
"step": 24000
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 6.47,
|
| 355 |
+
"learning_rate": 3.7595462713387242e-06,
|
| 356 |
+
"loss": 0.0793,
|
| 357 |
+
"step": 24500
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 6.6,
|
| 361 |
+
"learning_rate": 3.6191599281221925e-06,
|
| 362 |
+
"loss": 0.0697,
|
| 363 |
+
"step": 25000
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 6.73,
|
| 367 |
+
"learning_rate": 3.4787735849056607e-06,
|
| 368 |
+
"loss": 0.0667,
|
| 369 |
+
"step": 25500
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 6.86,
|
| 373 |
+
"learning_rate": 3.338387241689129e-06,
|
| 374 |
+
"loss": 0.0666,
|
| 375 |
+
"step": 26000
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 6.99,
|
| 379 |
+
"learning_rate": 3.1980008984725967e-06,
|
| 380 |
+
"loss": 0.0661,
|
| 381 |
+
"step": 26500
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 7.0,
|
| 385 |
+
"eval_accuracy": 0.9515961395694135,
|
| 386 |
+
"eval_loss": 0.32847702503204346,
|
| 387 |
+
"eval_runtime": 4.9266,
|
| 388 |
+
"eval_samples_per_second": 1367.065,
|
| 389 |
+
"eval_steps_per_second": 85.454,
|
| 390 |
+
"step": 26523
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 7.13,
|
| 394 |
+
"learning_rate": 3.057614555256065e-06,
|
| 395 |
+
"loss": 0.0561,
|
| 396 |
+
"step": 27000
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 7.26,
|
| 400 |
+
"learning_rate": 2.9172282120395328e-06,
|
| 401 |
+
"loss": 0.0596,
|
| 402 |
+
"step": 27500
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 7.39,
|
| 406 |
+
"learning_rate": 2.776841868823001e-06,
|
| 407 |
+
"loss": 0.053,
|
| 408 |
+
"step": 28000
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 7.52,
|
| 412 |
+
"learning_rate": 2.6364555256064692e-06,
|
| 413 |
+
"loss": 0.0546,
|
| 414 |
+
"step": 28500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 7.65,
|
| 418 |
+
"learning_rate": 2.496069182389937e-06,
|
| 419 |
+
"loss": 0.0559,
|
| 420 |
+
"step": 29000
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 7.79,
|
| 424 |
+
"learning_rate": 2.3556828391734053e-06,
|
| 425 |
+
"loss": 0.0686,
|
| 426 |
+
"step": 29500
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 7.92,
|
| 430 |
+
"learning_rate": 2.2152964959568735e-06,
|
| 431 |
+
"loss": 0.0632,
|
| 432 |
+
"step": 30000
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 8.0,
|
| 436 |
+
"eval_accuracy": 0.9504083147735709,
|
| 437 |
+
"eval_loss": 0.35768425464630127,
|
| 438 |
+
"eval_runtime": 4.9267,
|
| 439 |
+
"eval_samples_per_second": 1367.037,
|
| 440 |
+
"eval_steps_per_second": 85.452,
|
| 441 |
+
"step": 30312
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 8.05,
|
| 445 |
+
"learning_rate": 2.0749101527403413e-06,
|
| 446 |
+
"loss": 0.0457,
|
| 447 |
+
"step": 30500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 8.18,
|
| 451 |
+
"learning_rate": 1.9345238095238096e-06,
|
| 452 |
+
"loss": 0.0512,
|
| 453 |
+
"step": 31000
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 8.31,
|
| 457 |
+
"learning_rate": 1.7941374663072778e-06,
|
| 458 |
+
"loss": 0.0426,
|
| 459 |
+
"step": 31500
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 8.45,
|
| 463 |
+
"learning_rate": 1.6537511230907458e-06,
|
| 464 |
+
"loss": 0.0536,
|
| 465 |
+
"step": 32000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 8.58,
|
| 469 |
+
"learning_rate": 1.513364779874214e-06,
|
| 470 |
+
"loss": 0.0429,
|
| 471 |
+
"step": 32500
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 8.71,
|
| 475 |
+
"learning_rate": 1.372978436657682e-06,
|
| 476 |
+
"loss": 0.0431,
|
| 477 |
+
"step": 33000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 8.84,
|
| 481 |
+
"learning_rate": 1.2325920934411503e-06,
|
| 482 |
+
"loss": 0.0471,
|
| 483 |
+
"step": 33500
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 8.97,
|
| 487 |
+
"learning_rate": 1.092205750224618e-06,
|
| 488 |
+
"loss": 0.049,
|
| 489 |
+
"step": 34000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 9.0,
|
| 493 |
+
"eval_accuracy": 0.9505567928730512,
|
| 494 |
+
"eval_loss": 0.35209420323371887,
|
| 495 |
+
"eval_runtime": 4.9373,
|
| 496 |
+
"eval_samples_per_second": 1364.109,
|
| 497 |
+
"eval_steps_per_second": 85.269,
|
| 498 |
+
"step": 34101
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 9.11,
|
| 502 |
+
"learning_rate": 9.518194070080863e-07,
|
| 503 |
+
"loss": 0.0449,
|
| 504 |
+
"step": 34500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 9.24,
|
| 508 |
+
"learning_rate": 8.114330637915545e-07,
|
| 509 |
+
"loss": 0.0357,
|
| 510 |
+
"step": 35000
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 9.37,
|
| 514 |
+
"learning_rate": 6.710467205750225e-07,
|
| 515 |
+
"loss": 0.0494,
|
| 516 |
+
"step": 35500
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 9.5,
|
| 520 |
+
"learning_rate": 5.306603773584906e-07,
|
| 521 |
+
"loss": 0.0331,
|
| 522 |
+
"step": 36000
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 9.63,
|
| 526 |
+
"learning_rate": 3.902740341419587e-07,
|
| 527 |
+
"loss": 0.0367,
|
| 528 |
+
"step": 36500
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 9.77,
|
| 532 |
+
"learning_rate": 2.4988769092542675e-07,
|
| 533 |
+
"loss": 0.0395,
|
| 534 |
+
"step": 37000
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 9.9,
|
| 538 |
+
"learning_rate": 1.095013477088949e-07,
|
| 539 |
+
"loss": 0.0326,
|
| 540 |
+
"step": 37500
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 10.0,
|
| 544 |
+
"eval_accuracy": 0.949072011878248,
|
| 545 |
+
"eval_loss": 0.37647131085395813,
|
| 546 |
+
"eval_runtime": 4.935,
|
| 547 |
+
"eval_samples_per_second": 1364.743,
|
| 548 |
+
"eval_steps_per_second": 85.309,
|
| 549 |
+
"step": 37890
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 10.0,
|
| 553 |
+
"step": 37890,
|
| 554 |
+
"total_flos": 1.11573681913992e+16,
|
| 555 |
+
"train_loss": 0.11990570215349557,
|
| 556 |
+
"train_runtime": 1605.649,
|
| 557 |
+
"train_samples_per_second": 377.505,
|
| 558 |
+
"train_steps_per_second": 23.598
|
| 559 |
+
}
|
| 560 |
+
],
|
| 561 |
+
"logging_steps": 500,
|
| 562 |
+
"max_steps": 37890,
|
| 563 |
+
"num_train_epochs": 10,
|
| 564 |
+
"save_steps": 500,
|
| 565 |
+
"total_flos": 1.11573681913992e+16,
|
| 566 |
+
"trial_name": null,
|
| 567 |
+
"trial_params": null
|
| 568 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d0881ba1335231f65ee869f122b0458a34815e6aa47651abb934d37d7c5ed39
|
| 3 |
+
size 4091
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|