commit from chengluli
Browse files- config.json +27 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +432 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "./my_blue_storage/algebra_nation_roberta/roberta-large-finetuned/checkpoint-34500",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForCausalLM"
|
| 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": 1024,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 4096,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "roberta",
|
| 18 |
+
"num_attention_heads": 16,
|
| 19 |
+
"num_hidden_layers": 24,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.11.3",
|
| 24 |
+
"type_vocab_size": 1,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 50265
|
| 27 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57a632c11d41ec60b7526c6633b0abad95a4989f07791b4897623b794305f8d5
|
| 3 |
+
size 2843525721
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb47d3db0273971683b36b5c442921e06b97554947da801cc10608b1308bb287
|
| 3 |
+
size 1421785643
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f9e936362652542bf64919fa147bb42bbd309c6ffab88804087c0b18ab8da9e5
|
| 3 |
+
size 21579
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d7675be19d4c6929f233fff660f14bc6f9f55240cdb0e63de764e5c2386952a
|
| 3 |
+
size 623
|
trainer_state.json
ADDED
|
@@ -0,0 +1,432 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 19.96370235934664,
|
| 5 |
+
"global_step": 22000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.45,
|
| 12 |
+
"learning_rate": 1.9546279491833034e-05,
|
| 13 |
+
"loss": 0.834,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.91,
|
| 18 |
+
"learning_rate": 1.9092558983666063e-05,
|
| 19 |
+
"loss": 0.8279,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.0,
|
| 24 |
+
"eval_loss": 0.8352304100990295,
|
| 25 |
+
"eval_runtime": 300.0386,
|
| 26 |
+
"eval_samples_per_second": 338.6,
|
| 27 |
+
"eval_steps_per_second": 5.293,
|
| 28 |
+
"step": 1102
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 1.36,
|
| 32 |
+
"learning_rate": 1.8638838475499095e-05,
|
| 33 |
+
"loss": 0.8249,
|
| 34 |
+
"step": 1500
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 1.81,
|
| 38 |
+
"learning_rate": 1.8185117967332124e-05,
|
| 39 |
+
"loss": 0.8218,
|
| 40 |
+
"step": 2000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 2.0,
|
| 44 |
+
"eval_loss": 0.8333624005317688,
|
| 45 |
+
"eval_runtime": 299.0242,
|
| 46 |
+
"eval_samples_per_second": 339.748,
|
| 47 |
+
"eval_steps_per_second": 5.311,
|
| 48 |
+
"step": 2204
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 2.27,
|
| 52 |
+
"learning_rate": 1.7731397459165156e-05,
|
| 53 |
+
"loss": 0.8171,
|
| 54 |
+
"step": 2500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.72,
|
| 58 |
+
"learning_rate": 1.7277676950998185e-05,
|
| 59 |
+
"loss": 0.8138,
|
| 60 |
+
"step": 3000
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 3.0,
|
| 64 |
+
"eval_loss": 0.8279737234115601,
|
| 65 |
+
"eval_runtime": 299.8977,
|
| 66 |
+
"eval_samples_per_second": 338.759,
|
| 67 |
+
"eval_steps_per_second": 5.295,
|
| 68 |
+
"step": 3306
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 3.18,
|
| 72 |
+
"learning_rate": 1.6823956442831217e-05,
|
| 73 |
+
"loss": 0.8094,
|
| 74 |
+
"step": 3500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 3.63,
|
| 78 |
+
"learning_rate": 1.637023593466425e-05,
|
| 79 |
+
"loss": 0.8084,
|
| 80 |
+
"step": 4000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 4.0,
|
| 84 |
+
"eval_loss": 0.8279556035995483,
|
| 85 |
+
"eval_runtime": 305.7805,
|
| 86 |
+
"eval_samples_per_second": 332.242,
|
| 87 |
+
"eval_steps_per_second": 5.193,
|
| 88 |
+
"step": 4408
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 4.08,
|
| 92 |
+
"learning_rate": 1.5916515426497278e-05,
|
| 93 |
+
"loss": 0.8028,
|
| 94 |
+
"step": 4500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 4.54,
|
| 98 |
+
"learning_rate": 1.546279491833031e-05,
|
| 99 |
+
"loss": 0.7997,
|
| 100 |
+
"step": 5000
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 4.99,
|
| 104 |
+
"learning_rate": 1.500907441016334e-05,
|
| 105 |
+
"loss": 0.7975,
|
| 106 |
+
"step": 5500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 5.0,
|
| 110 |
+
"eval_loss": 0.8251700401306152,
|
| 111 |
+
"eval_runtime": 299.8544,
|
| 112 |
+
"eval_samples_per_second": 338.808,
|
| 113 |
+
"eval_steps_per_second": 5.296,
|
| 114 |
+
"step": 5510
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 5.44,
|
| 118 |
+
"learning_rate": 1.4555353901996372e-05,
|
| 119 |
+
"loss": 0.798,
|
| 120 |
+
"step": 6000
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 5.9,
|
| 124 |
+
"learning_rate": 1.4101633393829401e-05,
|
| 125 |
+
"loss": 0.7965,
|
| 126 |
+
"step": 6500
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 6.0,
|
| 130 |
+
"eval_loss": 0.8234975337982178,
|
| 131 |
+
"eval_runtime": 299.5265,
|
| 132 |
+
"eval_samples_per_second": 339.179,
|
| 133 |
+
"eval_steps_per_second": 5.302,
|
| 134 |
+
"step": 6612
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 6.35,
|
| 138 |
+
"learning_rate": 1.3647912885662433e-05,
|
| 139 |
+
"loss": 0.7913,
|
| 140 |
+
"step": 7000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 6.81,
|
| 144 |
+
"learning_rate": 1.3194192377495462e-05,
|
| 145 |
+
"loss": 0.7858,
|
| 146 |
+
"step": 7500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 7.0,
|
| 150 |
+
"eval_loss": 0.8199203014373779,
|
| 151 |
+
"eval_runtime": 298.6313,
|
| 152 |
+
"eval_samples_per_second": 340.195,
|
| 153 |
+
"eval_steps_per_second": 5.318,
|
| 154 |
+
"step": 7714
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 7.26,
|
| 158 |
+
"learning_rate": 1.2740471869328494e-05,
|
| 159 |
+
"loss": 0.7845,
|
| 160 |
+
"step": 8000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 7.71,
|
| 164 |
+
"learning_rate": 1.2286751361161527e-05,
|
| 165 |
+
"loss": 0.7808,
|
| 166 |
+
"step": 8500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 8.0,
|
| 170 |
+
"eval_loss": 0.8201999664306641,
|
| 171 |
+
"eval_runtime": 305.3068,
|
| 172 |
+
"eval_samples_per_second": 332.757,
|
| 173 |
+
"eval_steps_per_second": 5.201,
|
| 174 |
+
"step": 8816
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 8.17,
|
| 178 |
+
"learning_rate": 1.1833030852994555e-05,
|
| 179 |
+
"loss": 0.7824,
|
| 180 |
+
"step": 9000
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 8.62,
|
| 184 |
+
"learning_rate": 1.1379310344827587e-05,
|
| 185 |
+
"loss": 0.7805,
|
| 186 |
+
"step": 9500
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 9.0,
|
| 190 |
+
"eval_loss": 0.8179581165313721,
|
| 191 |
+
"eval_runtime": 299.8221,
|
| 192 |
+
"eval_samples_per_second": 338.844,
|
| 193 |
+
"eval_steps_per_second": 5.296,
|
| 194 |
+
"step": 9918
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 9.07,
|
| 198 |
+
"learning_rate": 1.0925589836660618e-05,
|
| 199 |
+
"loss": 0.7762,
|
| 200 |
+
"step": 10000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 9.53,
|
| 204 |
+
"learning_rate": 1.0471869328493648e-05,
|
| 205 |
+
"loss": 0.7744,
|
| 206 |
+
"step": 10500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 9.98,
|
| 210 |
+
"learning_rate": 1.0018148820326679e-05,
|
| 211 |
+
"loss": 0.7698,
|
| 212 |
+
"step": 11000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 10.0,
|
| 216 |
+
"eval_loss": 0.813572347164154,
|
| 217 |
+
"eval_runtime": 299.3306,
|
| 218 |
+
"eval_samples_per_second": 339.401,
|
| 219 |
+
"eval_steps_per_second": 5.305,
|
| 220 |
+
"step": 11020
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 10.44,
|
| 224 |
+
"learning_rate": 9.564428312159711e-06,
|
| 225 |
+
"loss": 0.768,
|
| 226 |
+
"step": 11500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 10.89,
|
| 230 |
+
"learning_rate": 9.110707803992742e-06,
|
| 231 |
+
"loss": 0.7704,
|
| 232 |
+
"step": 12000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 11.0,
|
| 236 |
+
"eval_loss": 0.8144657015800476,
|
| 237 |
+
"eval_runtime": 299.3629,
|
| 238 |
+
"eval_samples_per_second": 339.364,
|
| 239 |
+
"eval_steps_per_second": 5.305,
|
| 240 |
+
"step": 12122
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 11.34,
|
| 244 |
+
"learning_rate": 8.656987295825772e-06,
|
| 245 |
+
"loss": 0.7649,
|
| 246 |
+
"step": 12500
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 11.8,
|
| 250 |
+
"learning_rate": 8.203266787658803e-06,
|
| 251 |
+
"loss": 0.7666,
|
| 252 |
+
"step": 13000
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 12.0,
|
| 256 |
+
"eval_loss": 0.8124042749404907,
|
| 257 |
+
"eval_runtime": 306.0664,
|
| 258 |
+
"eval_samples_per_second": 331.931,
|
| 259 |
+
"eval_steps_per_second": 5.188,
|
| 260 |
+
"step": 13224
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 12.25,
|
| 264 |
+
"learning_rate": 7.749546279491835e-06,
|
| 265 |
+
"loss": 0.7638,
|
| 266 |
+
"step": 13500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 12.7,
|
| 270 |
+
"learning_rate": 7.295825771324865e-06,
|
| 271 |
+
"loss": 0.7616,
|
| 272 |
+
"step": 14000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 13.0,
|
| 276 |
+
"eval_loss": 0.8092049360275269,
|
| 277 |
+
"eval_runtime": 300.1177,
|
| 278 |
+
"eval_samples_per_second": 338.511,
|
| 279 |
+
"eval_steps_per_second": 5.291,
|
| 280 |
+
"step": 14326
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 13.16,
|
| 284 |
+
"learning_rate": 6.842105263157896e-06,
|
| 285 |
+
"loss": 0.7597,
|
| 286 |
+
"step": 14500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 13.61,
|
| 290 |
+
"learning_rate": 6.388384754990926e-06,
|
| 291 |
+
"loss": 0.7558,
|
| 292 |
+
"step": 15000
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 14.0,
|
| 296 |
+
"eval_loss": 0.8094133138656616,
|
| 297 |
+
"eval_runtime": 299.6821,
|
| 298 |
+
"eval_samples_per_second": 339.003,
|
| 299 |
+
"eval_steps_per_second": 5.299,
|
| 300 |
+
"step": 15428
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 14.07,
|
| 304 |
+
"learning_rate": 5.934664246823957e-06,
|
| 305 |
+
"loss": 0.7575,
|
| 306 |
+
"step": 15500
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 14.52,
|
| 310 |
+
"learning_rate": 5.480943738656987e-06,
|
| 311 |
+
"loss": 0.7562,
|
| 312 |
+
"step": 16000
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 14.97,
|
| 316 |
+
"learning_rate": 5.027223230490018e-06,
|
| 317 |
+
"loss": 0.7564,
|
| 318 |
+
"step": 16500
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 15.0,
|
| 322 |
+
"eval_loss": 0.8088417053222656,
|
| 323 |
+
"eval_runtime": 300.0855,
|
| 324 |
+
"eval_samples_per_second": 338.547,
|
| 325 |
+
"eval_steps_per_second": 5.292,
|
| 326 |
+
"step": 16530
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 15.43,
|
| 330 |
+
"learning_rate": 4.57350272232305e-06,
|
| 331 |
+
"loss": 0.7526,
|
| 332 |
+
"step": 17000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 15.88,
|
| 336 |
+
"learning_rate": 4.11978221415608e-06,
|
| 337 |
+
"loss": 0.753,
|
| 338 |
+
"step": 17500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 16.0,
|
| 342 |
+
"eval_loss": 0.8049781918525696,
|
| 343 |
+
"eval_runtime": 304.2347,
|
| 344 |
+
"eval_samples_per_second": 333.93,
|
| 345 |
+
"eval_steps_per_second": 5.22,
|
| 346 |
+
"step": 17632
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 16.33,
|
| 350 |
+
"learning_rate": 3.666061705989111e-06,
|
| 351 |
+
"loss": 0.7524,
|
| 352 |
+
"step": 18000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 16.79,
|
| 356 |
+
"learning_rate": 3.212341197822142e-06,
|
| 357 |
+
"loss": 0.7502,
|
| 358 |
+
"step": 18500
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 17.0,
|
| 362 |
+
"eval_loss": 0.8041799068450928,
|
| 363 |
+
"eval_runtime": 301.0417,
|
| 364 |
+
"eval_samples_per_second": 337.472,
|
| 365 |
+
"eval_steps_per_second": 5.275,
|
| 366 |
+
"step": 18734
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 17.24,
|
| 370 |
+
"learning_rate": 2.7586206896551725e-06,
|
| 371 |
+
"loss": 0.7487,
|
| 372 |
+
"step": 19000
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 17.7,
|
| 376 |
+
"learning_rate": 2.3049001814882034e-06,
|
| 377 |
+
"loss": 0.7495,
|
| 378 |
+
"step": 19500
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 18.0,
|
| 382 |
+
"eval_loss": 0.8050508499145508,
|
| 383 |
+
"eval_runtime": 298.8367,
|
| 384 |
+
"eval_samples_per_second": 339.962,
|
| 385 |
+
"eval_steps_per_second": 5.314,
|
| 386 |
+
"step": 19836
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 18.15,
|
| 390 |
+
"learning_rate": 1.8511796733212343e-06,
|
| 391 |
+
"loss": 0.7478,
|
| 392 |
+
"step": 20000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 18.6,
|
| 396 |
+
"learning_rate": 1.3974591651542652e-06,
|
| 397 |
+
"loss": 0.7479,
|
| 398 |
+
"step": 20500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 19.0,
|
| 402 |
+
"eval_loss": 0.8038392066955566,
|
| 403 |
+
"eval_runtime": 299.5234,
|
| 404 |
+
"eval_samples_per_second": 339.182,
|
| 405 |
+
"eval_steps_per_second": 5.302,
|
| 406 |
+
"step": 20938
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 19.06,
|
| 410 |
+
"learning_rate": 9.43738656987296e-07,
|
| 411 |
+
"loss": 0.7469,
|
| 412 |
+
"step": 21000
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 19.51,
|
| 416 |
+
"learning_rate": 4.900181488203267e-07,
|
| 417 |
+
"loss": 0.7473,
|
| 418 |
+
"step": 21500
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 19.96,
|
| 422 |
+
"learning_rate": 3.629764065335753e-08,
|
| 423 |
+
"loss": 0.7475,
|
| 424 |
+
"step": 22000
|
| 425 |
+
}
|
| 426 |
+
],
|
| 427 |
+
"max_steps": 22040,
|
| 428 |
+
"num_train_epochs": 20,
|
| 429 |
+
"total_flos": 2.622716924469212e+18,
|
| 430 |
+
"trial_name": null,
|
| 431 |
+
"trial_params": null
|
| 432 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a13d91b17249a3f242cf550317d2299e598c61a30e3d9905abd35f91ea0e4346
|
| 3 |
+
size 2927
|