commit from yhk
Browse files
checkpoint-17000/config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"gradient_checkpointing": false,
|
| 7 |
+
"hidden_act": "gelu",
|
| 8 |
+
"hidden_dropout_prob": 0.1,
|
| 9 |
+
"hidden_size": 768,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"intermediate_size": 3072,
|
| 12 |
+
"layer_norm_eps": 1e-12,
|
| 13 |
+
"max_position_embeddings": 514,
|
| 14 |
+
"model_type": "bert",
|
| 15 |
+
"num_attention_heads": 12,
|
| 16 |
+
"num_hidden_layers": 6,
|
| 17 |
+
"pad_token_id": 0,
|
| 18 |
+
"position_embedding_type": "absolute",
|
| 19 |
+
"transformers_version": "4.3.0",
|
| 20 |
+
"type_vocab_size": 1,
|
| 21 |
+
"use_cache": true,
|
| 22 |
+
"vocab_size": 52000
|
| 23 |
+
}
|
checkpoint-17000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:988592dd2bf22800026e01ab845cdc75a435bf476943de01b6fc6da8e8107bae
|
| 3 |
+
size 668100230
|
checkpoint-17000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70cb193d67a35a82de5827eca03840b6ccfa6fb95ff8acdbba867b5bfa99542c
|
| 3 |
+
size 334065777
|
checkpoint-17000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b60cd3fcc6c3ebe3dc2b872e74161969cc0ebbc018a72375e4bf11bf22cd1c6
|
| 3 |
+
size 623
|
checkpoint-17000/trainer_state.json
ADDED
|
@@ -0,0 +1,220 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 195.4022988505747,
|
| 5 |
+
"global_step": 17000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 5.75,
|
| 12 |
+
"learning_rate": 4.85632183908046e-05,
|
| 13 |
+
"loss": 7.2186,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 11.49,
|
| 18 |
+
"learning_rate": 4.7126436781609195e-05,
|
| 19 |
+
"loss": 6.4783,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 17.24,
|
| 24 |
+
"learning_rate": 4.5689655172413794e-05,
|
| 25 |
+
"loss": 6.2086,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 22.99,
|
| 30 |
+
"learning_rate": 4.4252873563218394e-05,
|
| 31 |
+
"loss": 5.8603,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 28.74,
|
| 36 |
+
"learning_rate": 4.2816091954022994e-05,
|
| 37 |
+
"loss": 5.5563,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 34.48,
|
| 42 |
+
"learning_rate": 4.1379310344827587e-05,
|
| 43 |
+
"loss": 5.311,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 40.23,
|
| 48 |
+
"learning_rate": 3.9942528735632186e-05,
|
| 49 |
+
"loss": 5.1382,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 45.98,
|
| 54 |
+
"learning_rate": 3.850574712643678e-05,
|
| 55 |
+
"loss": 4.9586,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 51.72,
|
| 60 |
+
"learning_rate": 3.7068965517241385e-05,
|
| 61 |
+
"loss": 4.7769,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 57.47,
|
| 66 |
+
"learning_rate": 3.563218390804598e-05,
|
| 67 |
+
"loss": 4.6488,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 63.22,
|
| 72 |
+
"learning_rate": 3.419540229885058e-05,
|
| 73 |
+
"loss": 4.5062,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 68.97,
|
| 78 |
+
"learning_rate": 3.275862068965517e-05,
|
| 79 |
+
"loss": 4.3904,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 74.71,
|
| 84 |
+
"learning_rate": 3.132183908045977e-05,
|
| 85 |
+
"loss": 4.2703,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 80.46,
|
| 90 |
+
"learning_rate": 2.988505747126437e-05,
|
| 91 |
+
"loss": 4.1789,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 86.21,
|
| 96 |
+
"learning_rate": 2.844827586206897e-05,
|
| 97 |
+
"loss": 4.0576,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 91.95,
|
| 102 |
+
"learning_rate": 2.7011494252873566e-05,
|
| 103 |
+
"loss": 3.9654,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 97.7,
|
| 108 |
+
"learning_rate": 2.5574712643678162e-05,
|
| 109 |
+
"loss": 3.8836,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 103.45,
|
| 114 |
+
"learning_rate": 2.413793103448276e-05,
|
| 115 |
+
"loss": 3.7926,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 109.2,
|
| 120 |
+
"learning_rate": 2.2701149425287358e-05,
|
| 121 |
+
"loss": 3.7162,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 114.94,
|
| 126 |
+
"learning_rate": 2.1264367816091954e-05,
|
| 127 |
+
"loss": 3.6335,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 120.69,
|
| 132 |
+
"learning_rate": 1.9827586206896554e-05,
|
| 133 |
+
"loss": 3.5557,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 126.44,
|
| 138 |
+
"learning_rate": 1.839080459770115e-05,
|
| 139 |
+
"loss": 3.5017,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 132.18,
|
| 144 |
+
"learning_rate": 1.6954022988505746e-05,
|
| 145 |
+
"loss": 3.4335,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 137.93,
|
| 150 |
+
"learning_rate": 1.5517241379310346e-05,
|
| 151 |
+
"loss": 3.3824,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 143.68,
|
| 156 |
+
"learning_rate": 1.4080459770114942e-05,
|
| 157 |
+
"loss": 3.3263,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 149.43,
|
| 162 |
+
"learning_rate": 1.2643678160919542e-05,
|
| 163 |
+
"loss": 3.275,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 155.17,
|
| 168 |
+
"learning_rate": 1.1206896551724138e-05,
|
| 169 |
+
"loss": 3.2154,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 160.92,
|
| 174 |
+
"learning_rate": 9.770114942528738e-06,
|
| 175 |
+
"loss": 3.188,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 166.67,
|
| 180 |
+
"learning_rate": 8.333333333333334e-06,
|
| 181 |
+
"loss": 3.1603,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 172.41,
|
| 186 |
+
"learning_rate": 6.896551724137932e-06,
|
| 187 |
+
"loss": 3.1172,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 178.16,
|
| 192 |
+
"learning_rate": 5.45977011494253e-06,
|
| 193 |
+
"loss": 3.0875,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 183.91,
|
| 198 |
+
"learning_rate": 4.022988505747127e-06,
|
| 199 |
+
"loss": 3.0623,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 189.66,
|
| 204 |
+
"learning_rate": 2.586206896551724e-06,
|
| 205 |
+
"loss": 3.0531,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 195.4,
|
| 210 |
+
"learning_rate": 1.1494252873563219e-06,
|
| 211 |
+
"loss": 3.0387,
|
| 212 |
+
"step": 17000
|
| 213 |
+
}
|
| 214 |
+
],
|
| 215 |
+
"max_steps": 17400,
|
| 216 |
+
"num_train_epochs": 200,
|
| 217 |
+
"total_flos": 86232908764058112,
|
| 218 |
+
"trial_name": null,
|
| 219 |
+
"trial_params": null
|
| 220 |
+
}
|
checkpoint-17000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c355ebc6fd3733b5b87f163750eb43fc76a681ce0015a60dd70faec1cdc3ac6c
|
| 3 |
+
size 2095
|