Upload checkpoint-63/trainer_state.json with huggingface_hub
Browse files- checkpoint-63/trainer_state.json +601 -0
checkpoint-63/trainer_state.json
ADDED
|
@@ -0,0 +1,601 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 63,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.016,
|
| 14 |
+
"grad_norm": 5.178493499755859,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 1.9778,
|
| 17 |
+
"mean_token_accuracy": 0.5731211602687836,
|
| 18 |
+
"num_tokens": 18095.0,
|
| 19 |
+
"step": 1
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.032,
|
| 23 |
+
"grad_norm": 4.289798736572266,
|
| 24 |
+
"learning_rate": 0.0001,
|
| 25 |
+
"loss": 2.0501,
|
| 26 |
+
"mean_token_accuracy": 0.5611206591129303,
|
| 27 |
+
"num_tokens": 33560.0,
|
| 28 |
+
"step": 2
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.048,
|
| 32 |
+
"grad_norm": 2.592556953430176,
|
| 33 |
+
"learning_rate": 0.0002,
|
| 34 |
+
"loss": 1.8085,
|
| 35 |
+
"mean_token_accuracy": 0.5969144999980927,
|
| 36 |
+
"num_tokens": 51416.0,
|
| 37 |
+
"step": 3
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.064,
|
| 41 |
+
"grad_norm": 1.5414831638336182,
|
| 42 |
+
"learning_rate": 0.00019988066808963474,
|
| 43 |
+
"loss": 1.825,
|
| 44 |
+
"mean_token_accuracy": 0.5979213863611221,
|
| 45 |
+
"num_tokens": 66606.0,
|
| 46 |
+
"step": 4
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.08,
|
| 50 |
+
"grad_norm": 1.532880425453186,
|
| 51 |
+
"learning_rate": 0.000199522988805313,
|
| 52 |
+
"loss": 1.6023,
|
| 53 |
+
"mean_token_accuracy": 0.6211032420396805,
|
| 54 |
+
"num_tokens": 81922.0,
|
| 55 |
+
"step": 5
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.096,
|
| 59 |
+
"grad_norm": 1.2121299505233765,
|
| 60 |
+
"learning_rate": 0.00019892791064819693,
|
| 61 |
+
"loss": 1.5619,
|
| 62 |
+
"mean_token_accuracy": 0.6331981718540192,
|
| 63 |
+
"num_tokens": 98973.0,
|
| 64 |
+
"step": 6
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.112,
|
| 68 |
+
"grad_norm": 0.9985714554786682,
|
| 69 |
+
"learning_rate": 0.00019809701165858222,
|
| 70 |
+
"loss": 1.401,
|
| 71 |
+
"mean_token_accuracy": 0.6640438586473465,
|
| 72 |
+
"num_tokens": 116786.0,
|
| 73 |
+
"step": 7
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.128,
|
| 77 |
+
"grad_norm": 1.235257863998413,
|
| 78 |
+
"learning_rate": 0.00019703249523121886,
|
| 79 |
+
"loss": 1.4001,
|
| 80 |
+
"mean_token_accuracy": 0.6453036367893219,
|
| 81 |
+
"num_tokens": 133738.0,
|
| 82 |
+
"step": 8
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.144,
|
| 86 |
+
"grad_norm": 1.1400103569030762,
|
| 87 |
+
"learning_rate": 0.00019573718427230442,
|
| 88 |
+
"loss": 1.2082,
|
| 89 |
+
"mean_token_accuracy": 0.6753191202878952,
|
| 90 |
+
"num_tokens": 155466.0,
|
| 91 |
+
"step": 9
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.16,
|
| 95 |
+
"grad_norm": 1.1436022520065308,
|
| 96 |
+
"learning_rate": 0.00019421451371364444,
|
| 97 |
+
"loss": 1.3161,
|
| 98 |
+
"mean_token_accuracy": 0.6673162132501602,
|
| 99 |
+
"num_tokens": 171140.0,
|
| 100 |
+
"step": 10
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.176,
|
| 104 |
+
"grad_norm": 0.9463217258453369,
|
| 105 |
+
"learning_rate": 0.00019246852140383043,
|
| 106 |
+
"loss": 1.3072,
|
| 107 |
+
"mean_token_accuracy": 0.662985697388649,
|
| 108 |
+
"num_tokens": 190013.0,
|
| 109 |
+
"step": 11
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.192,
|
| 113 |
+
"grad_norm": 0.718537449836731,
|
| 114 |
+
"learning_rate": 0.00019050383740059162,
|
| 115 |
+
"loss": 1.2311,
|
| 116 |
+
"mean_token_accuracy": 0.6756905913352966,
|
| 117 |
+
"num_tokens": 204990.0,
|
| 118 |
+
"step": 12
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.208,
|
| 122 |
+
"grad_norm": 0.5802047252655029,
|
| 123 |
+
"learning_rate": 0.000188325671692714,
|
| 124 |
+
"loss": 1.1599,
|
| 125 |
+
"mean_token_accuracy": 0.6960069835186005,
|
| 126 |
+
"num_tokens": 221669.0,
|
| 127 |
+
"step": 13
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.224,
|
| 131 |
+
"grad_norm": 0.6142511367797852,
|
| 132 |
+
"learning_rate": 0.0001859398003840867,
|
| 133 |
+
"loss": 1.1385,
|
| 134 |
+
"mean_token_accuracy": 0.6940146088600159,
|
| 135 |
+
"num_tokens": 238093.0,
|
| 136 |
+
"step": 14
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.24,
|
| 140 |
+
"grad_norm": 0.6557261347770691,
|
| 141 |
+
"learning_rate": 0.00018335255037651302,
|
| 142 |
+
"loss": 1.2242,
|
| 143 |
+
"mean_token_accuracy": 0.676676332950592,
|
| 144 |
+
"num_tokens": 255351.0,
|
| 145 |
+
"step": 15
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.256,
|
| 149 |
+
"grad_norm": 0.6320156455039978,
|
| 150 |
+
"learning_rate": 0.00018057078259190397,
|
| 151 |
+
"loss": 1.1493,
|
| 152 |
+
"mean_token_accuracy": 0.6945444643497467,
|
| 153 |
+
"num_tokens": 272261.0,
|
| 154 |
+
"step": 16
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.272,
|
| 158 |
+
"grad_norm": 0.58851158618927,
|
| 159 |
+
"learning_rate": 0.0001776018737783468,
|
| 160 |
+
"loss": 1.2255,
|
| 161 |
+
"mean_token_accuracy": 0.681264117360115,
|
| 162 |
+
"num_tokens": 288517.0,
|
| 163 |
+
"step": 17
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.288,
|
| 167 |
+
"grad_norm": 0.5777610540390015,
|
| 168 |
+
"learning_rate": 0.0001744536969482954,
|
| 169 |
+
"loss": 1.1277,
|
| 170 |
+
"mean_token_accuracy": 0.7005439400672913,
|
| 171 |
+
"num_tokens": 303397.0,
|
| 172 |
+
"step": 18
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.304,
|
| 176 |
+
"grad_norm": 0.4685754179954529,
|
| 177 |
+
"learning_rate": 0.00017113460050075638,
|
| 178 |
+
"loss": 1.0902,
|
| 179 |
+
"mean_token_accuracy": 0.7063727080821991,
|
| 180 |
+
"num_tokens": 320794.0,
|
| 181 |
+
"step": 19
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.32,
|
| 185 |
+
"grad_norm": 0.4389488101005554,
|
| 186 |
+
"learning_rate": 0.0001676533860828358,
|
| 187 |
+
"loss": 1.1166,
|
| 188 |
+
"mean_token_accuracy": 0.7016228437423706,
|
| 189 |
+
"num_tokens": 335928.0,
|
| 190 |
+
"step": 20
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.336,
|
| 194 |
+
"grad_norm": 0.47058454155921936,
|
| 195 |
+
"learning_rate": 0.00016401928524935314,
|
| 196 |
+
"loss": 1.0497,
|
| 197 |
+
"mean_token_accuracy": 0.7144980132579803,
|
| 198 |
+
"num_tokens": 353143.0,
|
| 199 |
+
"step": 21
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.352,
|
| 203 |
+
"grad_norm": 0.444322794675827,
|
| 204 |
+
"learning_rate": 0.0001602419349824178,
|
| 205 |
+
"loss": 1.0521,
|
| 206 |
+
"mean_token_accuracy": 0.7101306021213531,
|
| 207 |
+
"num_tokens": 371702.0,
|
| 208 |
+
"step": 22
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.368,
|
| 212 |
+
"grad_norm": 0.3345094621181488,
|
| 213 |
+
"learning_rate": 0.0001563313521358848,
|
| 214 |
+
"loss": 1.0394,
|
| 215 |
+
"mean_token_accuracy": 0.7164563238620758,
|
| 216 |
+
"num_tokens": 390548.0,
|
| 217 |
+
"step": 23
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.384,
|
| 221 |
+
"grad_norm": 0.318149209022522,
|
| 222 |
+
"learning_rate": 0.00015229790687245882,
|
| 223 |
+
"loss": 1.0316,
|
| 224 |
+
"mean_token_accuracy": 0.7225096970796585,
|
| 225 |
+
"num_tokens": 408577.0,
|
| 226 |
+
"step": 24
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.4,
|
| 230 |
+
"grad_norm": 0.36164650321006775,
|
| 231 |
+
"learning_rate": 0.0001481522951638875,
|
| 232 |
+
"loss": 1.0928,
|
| 233 |
+
"mean_token_accuracy": 0.7029632329940796,
|
| 234 |
+
"num_tokens": 426544.0,
|
| 235 |
+
"step": 25
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.416,
|
| 239 |
+
"grad_norm": 0.3186567723751068,
|
| 240 |
+
"learning_rate": 0.00014390551042716668,
|
| 241 |
+
"loss": 1.0932,
|
| 242 |
+
"mean_token_accuracy": 0.7054416686296463,
|
| 243 |
+
"num_tokens": 444668.0,
|
| 244 |
+
"step": 26
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 0.432,
|
| 248 |
+
"grad_norm": 0.3198276460170746,
|
| 249 |
+
"learning_rate": 0.00013956881437197514,
|
| 250 |
+
"loss": 0.8921,
|
| 251 |
+
"mean_token_accuracy": 0.7491990774869919,
|
| 252 |
+
"num_tokens": 460783.0,
|
| 253 |
+
"step": 27
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.448,
|
| 257 |
+
"grad_norm": 0.3033614158630371,
|
| 258 |
+
"learning_rate": 0.00013515370713664487,
|
| 259 |
+
"loss": 1.0343,
|
| 260 |
+
"mean_token_accuracy": 0.7144427299499512,
|
| 261 |
+
"num_tokens": 479754.0,
|
| 262 |
+
"step": 28
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.464,
|
| 266 |
+
"grad_norm": 0.3438175618648529,
|
| 267 |
+
"learning_rate": 0.00013067189679186162,
|
| 268 |
+
"loss": 1.0337,
|
| 269 |
+
"mean_token_accuracy": 0.7172342389822006,
|
| 270 |
+
"num_tokens": 496210.0,
|
| 271 |
+
"step": 29
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.48,
|
| 275 |
+
"grad_norm": 0.31151092052459717,
|
| 276 |
+
"learning_rate": 0.00012613526829296622,
|
| 277 |
+
"loss": 1.0293,
|
| 278 |
+
"mean_token_accuracy": 0.7132327109575272,
|
| 279 |
+
"num_tokens": 513602.0,
|
| 280 |
+
"step": 30
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.496,
|
| 284 |
+
"grad_norm": 0.37670907378196716,
|
| 285 |
+
"learning_rate": 0.0001215558519631896,
|
| 286 |
+
"loss": 1.0391,
|
| 287 |
+
"mean_token_accuracy": 0.7135330736637115,
|
| 288 |
+
"num_tokens": 528716.0,
|
| 289 |
+
"step": 31
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.512,
|
| 293 |
+
"grad_norm": 0.332743376493454,
|
| 294 |
+
"learning_rate": 0.0001169457915913982,
|
| 295 |
+
"loss": 1.0581,
|
| 296 |
+
"mean_token_accuracy": 0.7063487768173218,
|
| 297 |
+
"num_tokens": 545676.0,
|
| 298 |
+
"step": 32
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.528,
|
| 302 |
+
"grad_norm": 0.33181434869766235,
|
| 303 |
+
"learning_rate": 0.000112317312228949,
|
| 304 |
+
"loss": 1.0115,
|
| 305 |
+
"mean_token_accuracy": 0.7126458883285522,
|
| 306 |
+
"num_tokens": 558919.0,
|
| 307 |
+
"step": 33
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 0.544,
|
| 311 |
+
"grad_norm": 0.29894986748695374,
|
| 312 |
+
"learning_rate": 0.00010768268777105104,
|
| 313 |
+
"loss": 1.1125,
|
| 314 |
+
"mean_token_accuracy": 0.7048945128917694,
|
| 315 |
+
"num_tokens": 575928.0,
|
| 316 |
+
"step": 34
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.56,
|
| 320 |
+
"grad_norm": 0.3128003776073456,
|
| 321 |
+
"learning_rate": 0.00010305420840860182,
|
| 322 |
+
"loss": 1.0834,
|
| 323 |
+
"mean_token_accuracy": 0.7064289450645447,
|
| 324 |
+
"num_tokens": 591942.0,
|
| 325 |
+
"step": 35
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.576,
|
| 329 |
+
"grad_norm": 0.29789265990257263,
|
| 330 |
+
"learning_rate": 9.844414803681041e-05,
|
| 331 |
+
"loss": 0.942,
|
| 332 |
+
"mean_token_accuracy": 0.7331829965114594,
|
| 333 |
+
"num_tokens": 607626.0,
|
| 334 |
+
"step": 36
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 0.592,
|
| 338 |
+
"grad_norm": 0.2763383686542511,
|
| 339 |
+
"learning_rate": 9.386473170703382e-05,
|
| 340 |
+
"loss": 1.0408,
|
| 341 |
+
"mean_token_accuracy": 0.705119863152504,
|
| 342 |
+
"num_tokens": 626687.0,
|
| 343 |
+
"step": 37
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 0.608,
|
| 347 |
+
"grad_norm": 0.27845677733421326,
|
| 348 |
+
"learning_rate": 8.932810320813843e-05,
|
| 349 |
+
"loss": 0.963,
|
| 350 |
+
"mean_token_accuracy": 0.7295580208301544,
|
| 351 |
+
"num_tokens": 643989.0,
|
| 352 |
+
"step": 38
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.624,
|
| 356 |
+
"grad_norm": 0.2680862247943878,
|
| 357 |
+
"learning_rate": 8.484629286335517e-05,
|
| 358 |
+
"loss": 1.0114,
|
| 359 |
+
"mean_token_accuracy": 0.7323517799377441,
|
| 360 |
+
"num_tokens": 663165.0,
|
| 361 |
+
"step": 39
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.64,
|
| 365 |
+
"grad_norm": 0.2760467231273651,
|
| 366 |
+
"learning_rate": 8.043118562802488e-05,
|
| 367 |
+
"loss": 0.9966,
|
| 368 |
+
"mean_token_accuracy": 0.7186093181371689,
|
| 369 |
+
"num_tokens": 681802.0,
|
| 370 |
+
"step": 40
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.656,
|
| 374 |
+
"grad_norm": 0.26565995812416077,
|
| 375 |
+
"learning_rate": 7.609448957283334e-05,
|
| 376 |
+
"loss": 0.9775,
|
| 377 |
+
"mean_token_accuracy": 0.7272415161132812,
|
| 378 |
+
"num_tokens": 699772.0,
|
| 379 |
+
"step": 41
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 0.672,
|
| 383 |
+
"grad_norm": 0.29315948486328125,
|
| 384 |
+
"learning_rate": 7.184770483611256e-05,
|
| 385 |
+
"loss": 0.8769,
|
| 386 |
+
"mean_token_accuracy": 0.7540310174226761,
|
| 387 |
+
"num_tokens": 715209.0,
|
| 388 |
+
"step": 42
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.688,
|
| 392 |
+
"grad_norm": 0.24506399035453796,
|
| 393 |
+
"learning_rate": 6.770209312754124e-05,
|
| 394 |
+
"loss": 0.9194,
|
| 395 |
+
"mean_token_accuracy": 0.74180668592453,
|
| 396 |
+
"num_tokens": 735810.0,
|
| 397 |
+
"step": 43
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 0.704,
|
| 401 |
+
"grad_norm": 0.3123375177383423,
|
| 402 |
+
"learning_rate": 6.366864786411526e-05,
|
| 403 |
+
"loss": 1.0908,
|
| 404 |
+
"mean_token_accuracy": 0.7076454162597656,
|
| 405 |
+
"num_tokens": 752738.0,
|
| 406 |
+
"step": 44
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.72,
|
| 410 |
+
"grad_norm": 0.2974242866039276,
|
| 411 |
+
"learning_rate": 5.9758065017582185e-05,
|
| 412 |
+
"loss": 1.044,
|
| 413 |
+
"mean_token_accuracy": 0.712090015411377,
|
| 414 |
+
"num_tokens": 766970.0,
|
| 415 |
+
"step": 45
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.736,
|
| 419 |
+
"grad_norm": 0.3142162561416626,
|
| 420 |
+
"learning_rate": 5.598071475064688e-05,
|
| 421 |
+
"loss": 1.1099,
|
| 422 |
+
"mean_token_accuracy": 0.7006032019853592,
|
| 423 |
+
"num_tokens": 781115.0,
|
| 424 |
+
"step": 46
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.752,
|
| 428 |
+
"grad_norm": 0.2571069598197937,
|
| 429 |
+
"learning_rate": 5.2346613917164246e-05,
|
| 430 |
+
"loss": 0.9873,
|
| 431 |
+
"mean_token_accuracy": 0.7231301814317703,
|
| 432 |
+
"num_tokens": 801039.0,
|
| 433 |
+
"step": 47
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.768,
|
| 437 |
+
"grad_norm": 0.2709988057613373,
|
| 438 |
+
"learning_rate": 4.886539949924362e-05,
|
| 439 |
+
"loss": 0.8684,
|
| 440 |
+
"mean_token_accuracy": 0.7532013207674026,
|
| 441 |
+
"num_tokens": 818783.0,
|
| 442 |
+
"step": 48
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.784,
|
| 446 |
+
"grad_norm": 0.2727779746055603,
|
| 447 |
+
"learning_rate": 4.554630305170462e-05,
|
| 448 |
+
"loss": 1.0595,
|
| 449 |
+
"mean_token_accuracy": 0.712963730096817,
|
| 450 |
+
"num_tokens": 836166.0,
|
| 451 |
+
"step": 49
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.8,
|
| 455 |
+
"grad_norm": 0.2671469449996948,
|
| 456 |
+
"learning_rate": 4.2398126221653236e-05,
|
| 457 |
+
"loss": 0.9982,
|
| 458 |
+
"mean_token_accuracy": 0.7242924720048904,
|
| 459 |
+
"num_tokens": 856465.0,
|
| 460 |
+
"step": 50
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.816,
|
| 464 |
+
"grad_norm": 0.29176750779151917,
|
| 465 |
+
"learning_rate": 3.9429217408096075e-05,
|
| 466 |
+
"loss": 0.9357,
|
| 467 |
+
"mean_token_accuracy": 0.7428720593452454,
|
| 468 |
+
"num_tokens": 874214.0,
|
| 469 |
+
"step": 51
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.832,
|
| 473 |
+
"grad_norm": 0.2942723035812378,
|
| 474 |
+
"learning_rate": 3.664744962348699e-05,
|
| 475 |
+
"loss": 1.0035,
|
| 476 |
+
"mean_token_accuracy": 0.7168596237897873,
|
| 477 |
+
"num_tokens": 890584.0,
|
| 478 |
+
"step": 52
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.848,
|
| 482 |
+
"grad_norm": 0.29683244228363037,
|
| 483 |
+
"learning_rate": 3.40601996159133e-05,
|
| 484 |
+
"loss": 1.0647,
|
| 485 |
+
"mean_token_accuracy": 0.7118570357561111,
|
| 486 |
+
"num_tokens": 906031.0,
|
| 487 |
+
"step": 53
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.864,
|
| 491 |
+
"grad_norm": 0.2509916126728058,
|
| 492 |
+
"learning_rate": 3.167432830728603e-05,
|
| 493 |
+
"loss": 0.9915,
|
| 494 |
+
"mean_token_accuracy": 0.7244430631399155,
|
| 495 |
+
"num_tokens": 926256.0,
|
| 496 |
+
"step": 54
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.88,
|
| 500 |
+
"grad_norm": 0.29129576683044434,
|
| 501 |
+
"learning_rate": 2.949616259940842e-05,
|
| 502 |
+
"loss": 1.0276,
|
| 503 |
+
"mean_token_accuracy": 0.712443009018898,
|
| 504 |
+
"num_tokens": 942399.0,
|
| 505 |
+
"step": 55
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.896,
|
| 509 |
+
"grad_norm": 0.27668559551239014,
|
| 510 |
+
"learning_rate": 2.7531478596169587e-05,
|
| 511 |
+
"loss": 1.0886,
|
| 512 |
+
"mean_token_accuracy": 0.7021819353103638,
|
| 513 |
+
"num_tokens": 959208.0,
|
| 514 |
+
"step": 56
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.912,
|
| 518 |
+
"grad_norm": 0.27250081300735474,
|
| 519 |
+
"learning_rate": 2.5785486286355586e-05,
|
| 520 |
+
"loss": 1.1176,
|
| 521 |
+
"mean_token_accuracy": 0.6994952410459518,
|
| 522 |
+
"num_tokens": 978283.0,
|
| 523 |
+
"step": 57
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 0.928,
|
| 527 |
+
"grad_norm": 0.47203993797302246,
|
| 528 |
+
"learning_rate": 2.4262815727695575e-05,
|
| 529 |
+
"loss": 1.0219,
|
| 530 |
+
"mean_token_accuracy": 0.7143173664808273,
|
| 531 |
+
"num_tokens": 991015.0,
|
| 532 |
+
"step": 58
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 0.944,
|
| 536 |
+
"grad_norm": 0.27971458435058594,
|
| 537 |
+
"learning_rate": 2.2967504768781168e-05,
|
| 538 |
+
"loss": 1.0169,
|
| 539 |
+
"mean_token_accuracy": 0.7231282144784927,
|
| 540 |
+
"num_tokens": 1008413.0,
|
| 541 |
+
"step": 59
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.96,
|
| 545 |
+
"grad_norm": 0.2779776155948639,
|
| 546 |
+
"learning_rate": 2.19029883414178e-05,
|
| 547 |
+
"loss": 0.9886,
|
| 548 |
+
"mean_token_accuracy": 0.721703976392746,
|
| 549 |
+
"num_tokens": 1025326.0,
|
| 550 |
+
"step": 60
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.976,
|
| 554 |
+
"grad_norm": 0.2798386812210083,
|
| 555 |
+
"learning_rate": 2.1072089351803067e-05,
|
| 556 |
+
"loss": 0.9339,
|
| 557 |
+
"mean_token_accuracy": 0.7390532195568085,
|
| 558 |
+
"num_tokens": 1041871.0,
|
| 559 |
+
"step": 61
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 0.992,
|
| 563 |
+
"grad_norm": 0.27825894951820374,
|
| 564 |
+
"learning_rate": 2.0477011194686986e-05,
|
| 565 |
+
"loss": 0.9924,
|
| 566 |
+
"mean_token_accuracy": 0.7242806553840637,
|
| 567 |
+
"num_tokens": 1058238.0,
|
| 568 |
+
"step": 62
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 1.0,
|
| 572 |
+
"grad_norm": 0.4403225779533386,
|
| 573 |
+
"learning_rate": 2.0119331910365256e-05,
|
| 574 |
+
"loss": 1.0934,
|
| 575 |
+
"mean_token_accuracy": 0.6979949474334717,
|
| 576 |
+
"num_tokens": 1065193.0,
|
| 577 |
+
"step": 63
|
| 578 |
+
}
|
| 579 |
+
],
|
| 580 |
+
"logging_steps": 1,
|
| 581 |
+
"max_steps": 63,
|
| 582 |
+
"num_input_tokens_seen": 0,
|
| 583 |
+
"num_train_epochs": 1,
|
| 584 |
+
"save_steps": 500,
|
| 585 |
+
"stateful_callbacks": {
|
| 586 |
+
"TrainerControl": {
|
| 587 |
+
"args": {
|
| 588 |
+
"should_epoch_stop": false,
|
| 589 |
+
"should_evaluate": false,
|
| 590 |
+
"should_log": false,
|
| 591 |
+
"should_save": true,
|
| 592 |
+
"should_training_stop": true
|
| 593 |
+
},
|
| 594 |
+
"attributes": {}
|
| 595 |
+
}
|
| 596 |
+
},
|
| 597 |
+
"total_flos": 1.995467289064105e+17,
|
| 598 |
+
"train_batch_size": 4,
|
| 599 |
+
"trial_name": null,
|
| 600 |
+
"trial_params": null
|
| 601 |
+
}
|