Commit
•
c8c5844
1
Parent(s):
956a314
commit from turing
Browse files- README.md +1 -0
- checkpoint-20000/config.json +25 -0
- checkpoint-20000/optimizer.pt +3 -0
- checkpoint-20000/pytorch_model.bin +3 -0
- checkpoint-20000/rng_state.pth +3 -0
- checkpoint-20000/scheduler.pt +3 -0
- checkpoint-20000/trainer_state.json +256 -0
- checkpoint-20000/training_args.bin +3 -0
- checkpoint-30000/config.json +25 -0
- checkpoint-30000/optimizer.pt +3 -0
- checkpoint-30000/pytorch_model.bin +3 -0
- checkpoint-30000/rng_state.pth +3 -0
- checkpoint-30000/scheduler.pt +3 -0
- checkpoint-30000/trainer_state.json +376 -0
- checkpoint-30000/training_args.bin +3 -0
- config.json +25 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- training_args.bin +3 -0
- vocab.json +0 -0
README.md
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
RoBERTa model pretrained on the Brazilian Federal Official Gazette (200k instances).
|
checkpoint-20000/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"eos_token_id": 2,
|
| 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": 514,
|
| 16 |
+
"model_type": "roberta",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 6,
|
| 19 |
+
"pad_token_id": 1,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"transformers_version": "4.6.1",
|
| 22 |
+
"type_vocab_size": 1,
|
| 23 |
+
"use_cache": true,
|
| 24 |
+
"vocab_size": 50000
|
| 25 |
+
}
|
checkpoint-20000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:029e26d79084d6bd8514a27907ef7b46ceefc57b3796733a223b2c9d66fb9e90
|
| 3 |
+
size 655797923
|
checkpoint-20000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cbf766abeafa32494a4f8f13a43b9e230c5b0f5e1c544b107ac23c6195f751a2
|
| 3 |
+
size 327915507
|
checkpoint-20000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:44ec7a1b8619c1a34daada5bcc4ec627691c6d6f441d54e2e3316bd165f66556
|
| 3 |
+
size 15690
|
checkpoint-20000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa993dc5301fb42663e4742fc71288dc53c94e23a924a505cab343ed668bb7a4
|
| 3 |
+
size 623
|
checkpoint-20000/trainer_state.json
ADDED
|
@@ -0,0 +1,256 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 6.4,
|
| 5 |
+
"global_step": 20000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.16,
|
| 12 |
+
"learning_rate": 4.92e-05,
|
| 13 |
+
"loss": 6.308,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.32,
|
| 18 |
+
"learning_rate": 4.8400000000000004e-05,
|
| 19 |
+
"loss": 5.437,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.48,
|
| 24 |
+
"learning_rate": 4.76e-05,
|
| 25 |
+
"loss": 4.9197,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.64,
|
| 30 |
+
"learning_rate": 4.6800000000000006e-05,
|
| 31 |
+
"loss": 4.2132,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.8,
|
| 36 |
+
"learning_rate": 4.600000000000001e-05,
|
| 37 |
+
"loss": 3.5741,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.96,
|
| 42 |
+
"learning_rate": 4.52e-05,
|
| 43 |
+
"loss": 3.1896,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.12,
|
| 48 |
+
"learning_rate": 4.44e-05,
|
| 49 |
+
"loss": 2.9179,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.28,
|
| 54 |
+
"learning_rate": 4.36e-05,
|
| 55 |
+
"loss": 2.689,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.44,
|
| 60 |
+
"learning_rate": 4.2800000000000004e-05,
|
| 61 |
+
"loss": 2.5473,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.6,
|
| 66 |
+
"learning_rate": 4.2e-05,
|
| 67 |
+
"loss": 2.414,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.76,
|
| 72 |
+
"learning_rate": 4.12e-05,
|
| 73 |
+
"loss": 2.3057,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.92,
|
| 78 |
+
"learning_rate": 4.0400000000000006e-05,
|
| 79 |
+
"loss": 2.2348,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.08,
|
| 84 |
+
"learning_rate": 3.960000000000001e-05,
|
| 85 |
+
"loss": 2.1517,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 2.24,
|
| 90 |
+
"learning_rate": 3.88e-05,
|
| 91 |
+
"loss": 2.0636,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 2.4,
|
| 96 |
+
"learning_rate": 3.8e-05,
|
| 97 |
+
"loss": 2.0165,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 2.56,
|
| 102 |
+
"learning_rate": 3.72e-05,
|
| 103 |
+
"loss": 1.952,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.72,
|
| 108 |
+
"learning_rate": 3.6400000000000004e-05,
|
| 109 |
+
"loss": 1.9072,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 2.88,
|
| 114 |
+
"learning_rate": 3.56e-05,
|
| 115 |
+
"loss": 1.8658,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 3.04,
|
| 120 |
+
"learning_rate": 3.48e-05,
|
| 121 |
+
"loss": 1.8263,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 3.2,
|
| 126 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 127 |
+
"loss": 1.7846,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 3.36,
|
| 132 |
+
"learning_rate": 3.32e-05,
|
| 133 |
+
"loss": 1.7591,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.52,
|
| 138 |
+
"learning_rate": 3.24e-05,
|
| 139 |
+
"loss": 1.7168,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 3.68,
|
| 144 |
+
"learning_rate": 3.16e-05,
|
| 145 |
+
"loss": 1.6956,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 3.84,
|
| 150 |
+
"learning_rate": 3.08e-05,
|
| 151 |
+
"loss": 1.6747,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 4.0,
|
| 156 |
+
"learning_rate": 3e-05,
|
| 157 |
+
"loss": 1.6421,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 4.16,
|
| 162 |
+
"learning_rate": 2.9199999999999998e-05,
|
| 163 |
+
"loss": 1.6169,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 4.32,
|
| 168 |
+
"learning_rate": 2.84e-05,
|
| 169 |
+
"loss": 1.5991,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 4.48,
|
| 174 |
+
"learning_rate": 2.7600000000000003e-05,
|
| 175 |
+
"loss": 1.5645,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 4.64,
|
| 180 |
+
"learning_rate": 2.6800000000000004e-05,
|
| 181 |
+
"loss": 1.5588,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 4.8,
|
| 186 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 187 |
+
"loss": 1.5424,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 4.96,
|
| 192 |
+
"learning_rate": 2.5200000000000003e-05,
|
| 193 |
+
"loss": 1.532,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 5.12,
|
| 198 |
+
"learning_rate": 2.44e-05,
|
| 199 |
+
"loss": 1.5137,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 5.28,
|
| 204 |
+
"learning_rate": 2.36e-05,
|
| 205 |
+
"loss": 1.4885,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 5.44,
|
| 210 |
+
"learning_rate": 2.2800000000000002e-05,
|
| 211 |
+
"loss": 1.475,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 5.6,
|
| 216 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 217 |
+
"loss": 1.4793,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 5.76,
|
| 222 |
+
"learning_rate": 2.12e-05,
|
| 223 |
+
"loss": 1.4521,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 5.92,
|
| 228 |
+
"learning_rate": 2.04e-05,
|
| 229 |
+
"loss": 1.4481,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 6.08,
|
| 234 |
+
"learning_rate": 1.9600000000000002e-05,
|
| 235 |
+
"loss": 1.4376,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 6.24,
|
| 240 |
+
"learning_rate": 1.88e-05,
|
| 241 |
+
"loss": 1.4147,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 6.4,
|
| 246 |
+
"learning_rate": 1.8e-05,
|
| 247 |
+
"loss": 1.4153,
|
| 248 |
+
"step": 20000
|
| 249 |
+
}
|
| 250 |
+
],
|
| 251 |
+
"max_steps": 31250,
|
| 252 |
+
"num_train_epochs": 10,
|
| 253 |
+
"total_flos": 4.028813279232e+16,
|
| 254 |
+
"trial_name": null,
|
| 255 |
+
"trial_params": null
|
| 256 |
+
}
|
checkpoint-20000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f824eafc709d2e1f691aa49a59884a91b6199d85cf8ad621e262b2f0dbdd113a
|
| 3 |
+
size 2415
|
checkpoint-30000/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"eos_token_id": 2,
|
| 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": 514,
|
| 16 |
+
"model_type": "roberta",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 6,
|
| 19 |
+
"pad_token_id": 1,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"transformers_version": "4.6.1",
|
| 22 |
+
"type_vocab_size": 1,
|
| 23 |
+
"use_cache": true,
|
| 24 |
+
"vocab_size": 50000
|
| 25 |
+
}
|
checkpoint-30000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1057e5da745e03a9e7abbe962f31c9762cb31507a0683412322f792de216cfb0
|
| 3 |
+
size 655797923
|
checkpoint-30000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:337233f6a8d6ee10c305ecdeecf2c80cfb671c5aacca0977c7a8fdfc4e23c5b6
|
| 3 |
+
size 327915507
|
checkpoint-30000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1e63dd6551c4f0bf24ae0b2ee83830c03f05c59daa92825d72036c7592580b83
|
| 3 |
+
size 15690
|
checkpoint-30000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ec390ac2a3cfc0518e37c7e8670fbd4778da49343aa079fee40f9b3f2ee9f31a
|
| 3 |
+
size 623
|
checkpoint-30000/trainer_state.json
ADDED
|
@@ -0,0 +1,376 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.6,
|
| 5 |
+
"global_step": 30000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.16,
|
| 12 |
+
"learning_rate": 4.92e-05,
|
| 13 |
+
"loss": 6.308,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.32,
|
| 18 |
+
"learning_rate": 4.8400000000000004e-05,
|
| 19 |
+
"loss": 5.437,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.48,
|
| 24 |
+
"learning_rate": 4.76e-05,
|
| 25 |
+
"loss": 4.9197,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.64,
|
| 30 |
+
"learning_rate": 4.6800000000000006e-05,
|
| 31 |
+
"loss": 4.2132,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.8,
|
| 36 |
+
"learning_rate": 4.600000000000001e-05,
|
| 37 |
+
"loss": 3.5741,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.96,
|
| 42 |
+
"learning_rate": 4.52e-05,
|
| 43 |
+
"loss": 3.1896,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.12,
|
| 48 |
+
"learning_rate": 4.44e-05,
|
| 49 |
+
"loss": 2.9179,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.28,
|
| 54 |
+
"learning_rate": 4.36e-05,
|
| 55 |
+
"loss": 2.689,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.44,
|
| 60 |
+
"learning_rate": 4.2800000000000004e-05,
|
| 61 |
+
"loss": 2.5473,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.6,
|
| 66 |
+
"learning_rate": 4.2e-05,
|
| 67 |
+
"loss": 2.414,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.76,
|
| 72 |
+
"learning_rate": 4.12e-05,
|
| 73 |
+
"loss": 2.3057,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.92,
|
| 78 |
+
"learning_rate": 4.0400000000000006e-05,
|
| 79 |
+
"loss": 2.2348,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.08,
|
| 84 |
+
"learning_rate": 3.960000000000001e-05,
|
| 85 |
+
"loss": 2.1517,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 2.24,
|
| 90 |
+
"learning_rate": 3.88e-05,
|
| 91 |
+
"loss": 2.0636,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 2.4,
|
| 96 |
+
"learning_rate": 3.8e-05,
|
| 97 |
+
"loss": 2.0165,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 2.56,
|
| 102 |
+
"learning_rate": 3.72e-05,
|
| 103 |
+
"loss": 1.952,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.72,
|
| 108 |
+
"learning_rate": 3.6400000000000004e-05,
|
| 109 |
+
"loss": 1.9072,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 2.88,
|
| 114 |
+
"learning_rate": 3.56e-05,
|
| 115 |
+
"loss": 1.8658,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 3.04,
|
| 120 |
+
"learning_rate": 3.48e-05,
|
| 121 |
+
"loss": 1.8263,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 3.2,
|
| 126 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 127 |
+
"loss": 1.7846,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 3.36,
|
| 132 |
+
"learning_rate": 3.32e-05,
|
| 133 |
+
"loss": 1.7591,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.52,
|
| 138 |
+
"learning_rate": 3.24e-05,
|
| 139 |
+
"loss": 1.7168,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 3.68,
|
| 144 |
+
"learning_rate": 3.16e-05,
|
| 145 |
+
"loss": 1.6956,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 3.84,
|
| 150 |
+
"learning_rate": 3.08e-05,
|
| 151 |
+
"loss": 1.6747,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 4.0,
|
| 156 |
+
"learning_rate": 3e-05,
|
| 157 |
+
"loss": 1.6421,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 4.16,
|
| 162 |
+
"learning_rate": 2.9199999999999998e-05,
|
| 163 |
+
"loss": 1.6169,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 4.32,
|
| 168 |
+
"learning_rate": 2.84e-05,
|
| 169 |
+
"loss": 1.5991,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 4.48,
|
| 174 |
+
"learning_rate": 2.7600000000000003e-05,
|
| 175 |
+
"loss": 1.5645,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 4.64,
|
| 180 |
+
"learning_rate": 2.6800000000000004e-05,
|
| 181 |
+
"loss": 1.5588,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 4.8,
|
| 186 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 187 |
+
"loss": 1.5424,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 4.96,
|
| 192 |
+
"learning_rate": 2.5200000000000003e-05,
|
| 193 |
+
"loss": 1.532,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 5.12,
|
| 198 |
+
"learning_rate": 2.44e-05,
|
| 199 |
+
"loss": 1.5137,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 5.28,
|
| 204 |
+
"learning_rate": 2.36e-05,
|
| 205 |
+
"loss": 1.4885,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 5.44,
|
| 210 |
+
"learning_rate": 2.2800000000000002e-05,
|
| 211 |
+
"loss": 1.475,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 5.6,
|
| 216 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 217 |
+
"loss": 1.4793,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 5.76,
|
| 222 |
+
"learning_rate": 2.12e-05,
|
| 223 |
+
"loss": 1.4521,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 5.92,
|
| 228 |
+
"learning_rate": 2.04e-05,
|
| 229 |
+
"loss": 1.4481,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 6.08,
|
| 234 |
+
"learning_rate": 1.9600000000000002e-05,
|
| 235 |
+
"loss": 1.4376,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 6.24,
|
| 240 |
+
"learning_rate": 1.88e-05,
|
| 241 |
+
"loss": 1.4147,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 6.4,
|
| 246 |
+
"learning_rate": 1.8e-05,
|
| 247 |
+
"loss": 1.4153,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 6.56,
|
| 252 |
+
"learning_rate": 1.7199999999999998e-05,
|
| 253 |
+
"loss": 1.4011,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 6.72,
|
| 258 |
+
"learning_rate": 1.6400000000000002e-05,
|
| 259 |
+
"loss": 1.3889,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 6.88,
|
| 264 |
+
"learning_rate": 1.56e-05,
|
| 265 |
+
"loss": 1.3834,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 7.04,
|
| 270 |
+
"learning_rate": 1.48e-05,
|
| 271 |
+
"loss": 1.3765,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 7.2,
|
| 276 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 277 |
+
"loss": 1.3716,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 7.36,
|
| 282 |
+
"learning_rate": 1.32e-05,
|
| 283 |
+
"loss": 1.3623,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 7.52,
|
| 288 |
+
"learning_rate": 1.24e-05,
|
| 289 |
+
"loss": 1.3495,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 7.68,
|
| 294 |
+
"learning_rate": 1.16e-05,
|
| 295 |
+
"loss": 1.3539,
|
| 296 |
+
"step": 24000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 7.84,
|
| 300 |
+
"learning_rate": 1.08e-05,
|
| 301 |
+
"loss": 1.3432,
|
| 302 |
+
"step": 24500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 8.0,
|
| 306 |
+
"learning_rate": 1e-05,
|
| 307 |
+
"loss": 1.3416,
|
| 308 |
+
"step": 25000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 8.16,
|
| 312 |
+
"learning_rate": 9.2e-06,
|
| 313 |
+
"loss": 1.3253,
|
| 314 |
+
"step": 25500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 8.32,
|
| 318 |
+
"learning_rate": 8.400000000000001e-06,
|
| 319 |
+
"loss": 1.3203,
|
| 320 |
+
"step": 26000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 8.48,
|
| 324 |
+
"learning_rate": 7.6e-06,
|
| 325 |
+
"loss": 1.319,
|
| 326 |
+
"step": 26500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 8.64,
|
| 330 |
+
"learning_rate": 6.800000000000001e-06,
|
| 331 |
+
"loss": 1.3196,
|
| 332 |
+
"step": 27000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 8.8,
|
| 336 |
+
"learning_rate": 6e-06,
|
| 337 |
+
"loss": 1.3185,
|
| 338 |
+
"step": 27500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 8.96,
|
| 342 |
+
"learning_rate": 5.2e-06,
|
| 343 |
+
"loss": 1.3155,
|
| 344 |
+
"step": 28000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 9.12,
|
| 348 |
+
"learning_rate": 4.4e-06,
|
| 349 |
+
"loss": 1.3083,
|
| 350 |
+
"step": 28500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 9.28,
|
| 354 |
+
"learning_rate": 3.6e-06,
|
| 355 |
+
"loss": 1.2983,
|
| 356 |
+
"step": 29000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 9.44,
|
| 360 |
+
"learning_rate": 2.8000000000000003e-06,
|
| 361 |
+
"loss": 1.2991,
|
| 362 |
+
"step": 29500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 9.6,
|
| 366 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 367 |
+
"loss": 1.3022,
|
| 368 |
+
"step": 30000
|
| 369 |
+
}
|
| 370 |
+
],
|
| 371 |
+
"max_steps": 31250,
|
| 372 |
+
"num_train_epochs": 10,
|
| 373 |
+
"total_flos": 4.028813279232e+16,
|
| 374 |
+
"trial_name": null,
|
| 375 |
+
"trial_params": null
|
| 376 |
+
}
|
checkpoint-30000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f824eafc709d2e1f691aa49a59884a91b6199d85cf8ad621e262b2f0dbdd113a
|
| 3 |
+
size 2415
|
config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"eos_token_id": 2,
|
| 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": 514,
|
| 16 |
+
"model_type": "roberta",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 6,
|
| 19 |
+
"pad_token_id": 1,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"transformers_version": "4.6.1",
|
| 22 |
+
"type_vocab_size": 1,
|
| 23 |
+
"use_cache": true,
|
| 24 |
+
"vocab_size": 50000
|
| 25 |
+
}
|
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 sha256:2c949bd9ec5ebcbcce002b975cf61f6f3a06374c20ad0ef67d88a541011b2655
|
| 3 |
+
size 327915507
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f824eafc709d2e1f691aa49a59884a91b6199d85cf8ad621e262b2f0dbdd113a
|
| 3 |
+
size 2415
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|