Commit ·
75eb3dc
1
Parent(s): deb445e
init
Browse files- config.json +30 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- train_results.txt +3 -0
- trainer_state.json +442 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "junnyu/wobert_chinese_base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForCL"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"directionality": "bidi",
|
| 8 |
+
"gradient_checkpointing": false,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 512,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
+
"pad_token_id": 0,
|
| 20 |
+
"pooler_fc_size": 768,
|
| 21 |
+
"pooler_num_attention_heads": 12,
|
| 22 |
+
"pooler_num_fc_layers": 3,
|
| 23 |
+
"pooler_size_per_head": 128,
|
| 24 |
+
"pooler_type": "first_token_transform",
|
| 25 |
+
"position_embedding_type": "absolute",
|
| 26 |
+
"transformers_version": "4.2.1",
|
| 27 |
+
"type_vocab_size": 2,
|
| 28 |
+
"use_cache": true,
|
| 29 |
+
"vocab_size": 33586
|
| 30 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": null, "name_or_path": "junnyu/wobert_chinese_base", "do_basic_tokenize": true, "never_split": null}
|
train_results.txt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
epoch = 12.77
|
| 2 |
+
train_runtime = 11218.2271
|
| 3 |
+
train_samples_per_second = 3.12
|
trainer_state.json
ADDED
|
@@ -0,0 +1,442 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 12.769062385990514,
|
| 5 |
+
"global_step": 35000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.18,
|
| 12 |
+
"learning_rate": 3.9428571428571435e-05,
|
| 13 |
+
"loss": 0.7626,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.36,
|
| 18 |
+
"learning_rate": 3.885714285714286e-05,
|
| 19 |
+
"loss": 0.3395,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.55,
|
| 24 |
+
"learning_rate": 3.828571428571429e-05,
|
| 25 |
+
"loss": 0.1906,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.73,
|
| 30 |
+
"learning_rate": 3.771428571428572e-05,
|
| 31 |
+
"loss": 0.1287,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.91,
|
| 36 |
+
"learning_rate": 3.714285714285715e-05,
|
| 37 |
+
"loss": 0.1061,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.09,
|
| 42 |
+
"learning_rate": 3.6571428571428576e-05,
|
| 43 |
+
"loss": 0.0865,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.28,
|
| 48 |
+
"learning_rate": 3.6e-05,
|
| 49 |
+
"loss": 0.0646,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.46,
|
| 54 |
+
"learning_rate": 3.542857142857143e-05,
|
| 55 |
+
"loss": 0.0562,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.64,
|
| 60 |
+
"learning_rate": 3.485714285714286e-05,
|
| 61 |
+
"loss": 0.0685,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.82,
|
| 66 |
+
"learning_rate": 3.4285714285714284e-05,
|
| 67 |
+
"loss": 0.0481,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.01,
|
| 72 |
+
"learning_rate": 3.3714285714285716e-05,
|
| 73 |
+
"loss": 0.0445,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.19,
|
| 78 |
+
"learning_rate": 3.314285714285715e-05,
|
| 79 |
+
"loss": 0.0345,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.37,
|
| 84 |
+
"learning_rate": 3.257142857142857e-05,
|
| 85 |
+
"loss": 0.0353,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 2.55,
|
| 90 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 91 |
+
"loss": 0.0436,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 2.74,
|
| 96 |
+
"learning_rate": 3.142857142857143e-05,
|
| 97 |
+
"loss": 0.0337,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 2.92,
|
| 102 |
+
"learning_rate": 3.085714285714286e-05,
|
| 103 |
+
"loss": 0.0325,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 3.1,
|
| 108 |
+
"learning_rate": 3.0285714285714288e-05,
|
| 109 |
+
"loss": 0.0325,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 3.28,
|
| 114 |
+
"learning_rate": 2.9714285714285717e-05,
|
| 115 |
+
"loss": 0.0303,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 3.47,
|
| 120 |
+
"learning_rate": 2.9142857142857146e-05,
|
| 121 |
+
"loss": 0.0243,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 3.65,
|
| 126 |
+
"learning_rate": 2.8571428571428574e-05,
|
| 127 |
+
"loss": 0.0269,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 3.83,
|
| 132 |
+
"learning_rate": 2.8e-05,
|
| 133 |
+
"loss": 0.0252,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 4.01,
|
| 138 |
+
"learning_rate": 2.742857142857143e-05,
|
| 139 |
+
"loss": 0.0285,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 4.2,
|
| 144 |
+
"learning_rate": 2.6857142857142857e-05,
|
| 145 |
+
"loss": 0.0265,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 4.38,
|
| 150 |
+
"learning_rate": 2.628571428571429e-05,
|
| 151 |
+
"loss": 0.0197,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 4.56,
|
| 156 |
+
"learning_rate": 2.5714285714285718e-05,
|
| 157 |
+
"loss": 0.0253,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 4.74,
|
| 162 |
+
"learning_rate": 2.5142857142857143e-05,
|
| 163 |
+
"loss": 0.025,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 4.93,
|
| 168 |
+
"learning_rate": 2.4571428571428575e-05,
|
| 169 |
+
"loss": 0.0254,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 5.11,
|
| 174 |
+
"learning_rate": 2.4e-05,
|
| 175 |
+
"loss": 0.0235,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 5.29,
|
| 180 |
+
"learning_rate": 2.3428571428571433e-05,
|
| 181 |
+
"loss": 0.025,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 5.47,
|
| 186 |
+
"learning_rate": 2.2857142857142858e-05,
|
| 187 |
+
"loss": 0.0191,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 5.65,
|
| 192 |
+
"learning_rate": 2.2285714285714287e-05,
|
| 193 |
+
"loss": 0.0225,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 5.84,
|
| 198 |
+
"learning_rate": 2.1714285714285715e-05,
|
| 199 |
+
"loss": 0.0195,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 6.02,
|
| 204 |
+
"learning_rate": 2.1142857142857144e-05,
|
| 205 |
+
"loss": 0.017,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 6.2,
|
| 210 |
+
"learning_rate": 2.057142857142857e-05,
|
| 211 |
+
"loss": 0.0196,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 6.38,
|
| 216 |
+
"learning_rate": 2e-05,
|
| 217 |
+
"loss": 0.0174,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 6.57,
|
| 222 |
+
"learning_rate": 1.942857142857143e-05,
|
| 223 |
+
"loss": 0.0205,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 6.75,
|
| 228 |
+
"learning_rate": 1.885714285714286e-05,
|
| 229 |
+
"loss": 0.0223,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 6.93,
|
| 234 |
+
"learning_rate": 1.8285714285714288e-05,
|
| 235 |
+
"loss": 0.0181,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 7.11,
|
| 240 |
+
"learning_rate": 1.7714285714285717e-05,
|
| 241 |
+
"loss": 0.026,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 7.3,
|
| 246 |
+
"learning_rate": 1.7142857142857142e-05,
|
| 247 |
+
"loss": 0.0191,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 7.48,
|
| 252 |
+
"learning_rate": 1.6571428571428574e-05,
|
| 253 |
+
"loss": 0.0194,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 7.66,
|
| 258 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 259 |
+
"loss": 0.021,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 7.84,
|
| 264 |
+
"learning_rate": 1.542857142857143e-05,
|
| 265 |
+
"loss": 0.0166,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 8.03,
|
| 270 |
+
"learning_rate": 1.4857142857142858e-05,
|
| 271 |
+
"loss": 0.017,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 8.21,
|
| 276 |
+
"learning_rate": 1.4285714285714287e-05,
|
| 277 |
+
"loss": 0.0185,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 8.39,
|
| 282 |
+
"learning_rate": 1.3714285714285716e-05,
|
| 283 |
+
"loss": 0.0179,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 8.57,
|
| 288 |
+
"learning_rate": 1.3142857142857145e-05,
|
| 289 |
+
"loss": 0.0179,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 8.76,
|
| 294 |
+
"learning_rate": 1.2571428571428572e-05,
|
| 295 |
+
"loss": 0.0175,
|
| 296 |
+
"step": 24000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 8.94,
|
| 300 |
+
"learning_rate": 1.2e-05,
|
| 301 |
+
"loss": 0.0146,
|
| 302 |
+
"step": 24500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 9.12,
|
| 306 |
+
"learning_rate": 1.1428571428571429e-05,
|
| 307 |
+
"loss": 0.0223,
|
| 308 |
+
"step": 25000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 9.3,
|
| 312 |
+
"learning_rate": 1.0857142857142858e-05,
|
| 313 |
+
"loss": 0.0173,
|
| 314 |
+
"step": 25500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 9.49,
|
| 318 |
+
"learning_rate": 1.0285714285714285e-05,
|
| 319 |
+
"loss": 0.0214,
|
| 320 |
+
"step": 26000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 9.67,
|
| 324 |
+
"learning_rate": 9.714285714285715e-06,
|
| 325 |
+
"loss": 0.0185,
|
| 326 |
+
"step": 26500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 9.85,
|
| 330 |
+
"learning_rate": 9.142857142857144e-06,
|
| 331 |
+
"loss": 0.021,
|
| 332 |
+
"step": 27000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 10.03,
|
| 336 |
+
"learning_rate": 8.571428571428571e-06,
|
| 337 |
+
"loss": 0.0191,
|
| 338 |
+
"step": 27500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 10.22,
|
| 342 |
+
"learning_rate": 8.000000000000001e-06,
|
| 343 |
+
"loss": 0.0145,
|
| 344 |
+
"step": 28000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 10.4,
|
| 348 |
+
"learning_rate": 7.428571428571429e-06,
|
| 349 |
+
"loss": 0.0187,
|
| 350 |
+
"step": 28500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 10.58,
|
| 354 |
+
"learning_rate": 6.857142857142858e-06,
|
| 355 |
+
"loss": 0.0184,
|
| 356 |
+
"step": 29000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 10.76,
|
| 360 |
+
"learning_rate": 6.285714285714286e-06,
|
| 361 |
+
"loss": 0.0171,
|
| 362 |
+
"step": 29500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 10.94,
|
| 366 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 367 |
+
"loss": 0.0167,
|
| 368 |
+
"step": 30000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 11.13,
|
| 372 |
+
"learning_rate": 5.142857142857142e-06,
|
| 373 |
+
"loss": 0.0193,
|
| 374 |
+
"step": 30500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 11.31,
|
| 378 |
+
"learning_rate": 4.571428571428572e-06,
|
| 379 |
+
"loss": 0.0151,
|
| 380 |
+
"step": 31000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 11.49,
|
| 384 |
+
"learning_rate": 4.000000000000001e-06,
|
| 385 |
+
"loss": 0.0148,
|
| 386 |
+
"step": 31500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 11.67,
|
| 390 |
+
"learning_rate": 3.428571428571429e-06,
|
| 391 |
+
"loss": 0.0202,
|
| 392 |
+
"step": 32000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 11.86,
|
| 396 |
+
"learning_rate": 2.8571428571428573e-06,
|
| 397 |
+
"loss": 0.0155,
|
| 398 |
+
"step": 32500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 12.04,
|
| 402 |
+
"learning_rate": 2.285714285714286e-06,
|
| 403 |
+
"loss": 0.017,
|
| 404 |
+
"step": 33000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 12.22,
|
| 408 |
+
"learning_rate": 1.7142857142857145e-06,
|
| 409 |
+
"loss": 0.022,
|
| 410 |
+
"step": 33500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 12.4,
|
| 414 |
+
"learning_rate": 1.142857142857143e-06,
|
| 415 |
+
"loss": 0.0176,
|
| 416 |
+
"step": 34000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 12.59,
|
| 420 |
+
"learning_rate": 5.714285714285715e-07,
|
| 421 |
+
"loss": 0.0158,
|
| 422 |
+
"step": 34500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 12.77,
|
| 426 |
+
"learning_rate": 0.0,
|
| 427 |
+
"loss": 0.0116,
|
| 428 |
+
"step": 35000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 12.77,
|
| 432 |
+
"step": 35000,
|
| 433 |
+
"train_runtime": 11218.2271,
|
| 434 |
+
"train_samples_per_second": 3.12
|
| 435 |
+
}
|
| 436 |
+
],
|
| 437 |
+
"max_steps": 35000,
|
| 438 |
+
"num_train_epochs": 13,
|
| 439 |
+
"total_flos": 72522034090868736,
|
| 440 |
+
"trial_name": null,
|
| 441 |
+
"trial_params": null
|
| 442 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b7605b2cc91e8b7f01e5e90f8905571564e02a06d9aa1f9132188bae1e1d4122
|
| 3 |
+
size 2031
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|