Upload folder using huggingface_hub
Browse files- generation_config.json +13 -0
- trainer_state.json +566 -0
generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_k": 20,
|
| 11 |
+
"top_p": 0.95,
|
| 12 |
+
"transformers_version": "4.51.0"
|
| 13 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,566 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.99867197875166,
|
| 6 |
+
"eval_steps": 1000,
|
| 7 |
+
"global_step": 376,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0026560424966799467,
|
| 14 |
+
"grad_norm": 1402.69921875,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 52.5156,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.013280212483399735,
|
| 21 |
+
"grad_norm": 636.0113525390625,
|
| 22 |
+
"learning_rate": 3.1578947368421045e-05,
|
| 23 |
+
"loss": 32.8066,
|
| 24 |
+
"step": 5
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.02656042496679947,
|
| 28 |
+
"grad_norm": 475.96771240234375,
|
| 29 |
+
"learning_rate": 7.105263157894735e-05,
|
| 30 |
+
"loss": 8.907,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.0398406374501992,
|
| 35 |
+
"grad_norm": 109.1396713256836,
|
| 36 |
+
"learning_rate": 0.00011052631578947366,
|
| 37 |
+
"loss": 6.3063,
|
| 38 |
+
"step": 15
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.05312084993359894,
|
| 42 |
+
"grad_norm": 12.496488571166992,
|
| 43 |
+
"learning_rate": 0.00015,
|
| 44 |
+
"loss": 4.4115,
|
| 45 |
+
"step": 20
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.06640106241699867,
|
| 49 |
+
"grad_norm": 88.81600952148438,
|
| 50 |
+
"learning_rate": 0.0001894736842105263,
|
| 51 |
+
"loss": 3.3893,
|
| 52 |
+
"step": 25
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.0796812749003984,
|
| 56 |
+
"grad_norm": 107.62394714355469,
|
| 57 |
+
"learning_rate": 0.00022894736842105263,
|
| 58 |
+
"loss": 9.3801,
|
| 59 |
+
"step": 30
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.09296148738379814,
|
| 63 |
+
"grad_norm": 7.125447750091553,
|
| 64 |
+
"learning_rate": 0.0002684210526315789,
|
| 65 |
+
"loss": 4.3988,
|
| 66 |
+
"step": 35
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.10624169986719788,
|
| 70 |
+
"grad_norm": 15.962409973144531,
|
| 71 |
+
"learning_rate": 0.0002999935207538156,
|
| 72 |
+
"loss": 4.009,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.11952191235059761,
|
| 77 |
+
"grad_norm": 11.288817405700684,
|
| 78 |
+
"learning_rate": 0.00029976680590482896,
|
| 79 |
+
"loss": 3.0982,
|
| 80 |
+
"step": 45
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.13280212483399734,
|
| 84 |
+
"grad_norm": 9.365798950195312,
|
| 85 |
+
"learning_rate": 0.0002992166882710483,
|
| 86 |
+
"loss": 5.2771,
|
| 87 |
+
"step": 50
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.1460823373173971,
|
| 91 |
+
"grad_norm": 52.74184036254883,
|
| 92 |
+
"learning_rate": 0.0002983443557630634,
|
| 93 |
+
"loss": 6.7707,
|
| 94 |
+
"step": 55
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.1593625498007968,
|
| 98 |
+
"grad_norm": 5.0402116775512695,
|
| 99 |
+
"learning_rate": 0.00029715169207439795,
|
| 100 |
+
"loss": 4.2547,
|
| 101 |
+
"step": 60
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.17264276228419656,
|
| 105 |
+
"grad_norm": 5.552392482757568,
|
| 106 |
+
"learning_rate": 0.00029564127261390776,
|
| 107 |
+
"loss": 2.7717,
|
| 108 |
+
"step": 65
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.18592297476759628,
|
| 112 |
+
"grad_norm": 3.872722864151001,
|
| 113 |
+
"learning_rate": 0.0002938163589445058,
|
| 114 |
+
"loss": 2.6939,
|
| 115 |
+
"step": 70
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.199203187250996,
|
| 119 |
+
"grad_norm": 5.162952899932861,
|
| 120 |
+
"learning_rate": 0.0002916808917402228,
|
| 121 |
+
"loss": 2.6598,
|
| 122 |
+
"step": 75
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.21248339973439576,
|
| 126 |
+
"grad_norm": 4.01698637008667,
|
| 127 |
+
"learning_rate": 0.0002892394822768119,
|
| 128 |
+
"loss": 2.6768,
|
| 129 |
+
"step": 80
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.22576361221779548,
|
| 133 |
+
"grad_norm": 3.66325306892395,
|
| 134 |
+
"learning_rate": 0.0002864974024742725,
|
| 135 |
+
"loss": 2.6551,
|
| 136 |
+
"step": 85
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.23904382470119523,
|
| 140 |
+
"grad_norm": 3.5804343223571777,
|
| 141 |
+
"learning_rate": 0.00028346057351279474,
|
| 142 |
+
"loss": 2.6633,
|
| 143 |
+
"step": 90
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.25232403718459495,
|
| 147 |
+
"grad_norm": 3.087800979614258,
|
| 148 |
+
"learning_rate": 0.00028013555304670765,
|
| 149 |
+
"loss": 2.632,
|
| 150 |
+
"step": 95
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.2656042496679947,
|
| 154 |
+
"grad_norm": 3.0033605098724365,
|
| 155 |
+
"learning_rate": 0.00027652952104404045,
|
| 156 |
+
"loss": 2.5992,
|
| 157 |
+
"step": 100
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.2788844621513944,
|
| 161 |
+
"grad_norm": 2.973752021789551,
|
| 162 |
+
"learning_rate": 0.0002726502642822747,
|
| 163 |
+
"loss": 2.552,
|
| 164 |
+
"step": 105
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.2921646746347942,
|
| 168 |
+
"grad_norm": 2.6142959594726562,
|
| 169 |
+
"learning_rate": 0.00026850615953376744,
|
| 170 |
+
"loss": 2.582,
|
| 171 |
+
"step": 110
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.3054448871181939,
|
| 175 |
+
"grad_norm": 2.8691508769989014,
|
| 176 |
+
"learning_rate": 0.00026410615547715297,
|
| 177 |
+
"loss": 2.5102,
|
| 178 |
+
"step": 115
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.3187250996015936,
|
| 182 |
+
"grad_norm": 3.016436815261841,
|
| 183 |
+
"learning_rate": 0.00025945975337378515,
|
| 184 |
+
"loss": 2.5391,
|
| 185 |
+
"step": 120
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.33200531208499334,
|
| 189 |
+
"grad_norm": 2.747736930847168,
|
| 190 |
+
"learning_rate": 0.00025457698655094495,
|
| 191 |
+
"loss": 2.5082,
|
| 192 |
+
"step": 125
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.3452855245683931,
|
| 196 |
+
"grad_norm": 2.5107271671295166,
|
| 197 |
+
"learning_rate": 0.00024946839873611924,
|
| 198 |
+
"loss": 2.5054,
|
| 199 |
+
"step": 130
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.35856573705179284,
|
| 203 |
+
"grad_norm": 2.436136484146118,
|
| 204 |
+
"learning_rate": 0.00024414502128913227,
|
| 205 |
+
"loss": 2.4813,
|
| 206 |
+
"step": 135
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.37184594953519257,
|
| 210 |
+
"grad_norm": 2.58207631111145,
|
| 211 |
+
"learning_rate": 0.00023861834938129696,
|
| 212 |
+
"loss": 2.4547,
|
| 213 |
+
"step": 140
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.3851261620185923,
|
| 217 |
+
"grad_norm": 2.4167120456695557,
|
| 218 |
+
"learning_rate": 0.00023290031717302254,
|
| 219 |
+
"loss": 2.4403,
|
| 220 |
+
"step": 145
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.398406374501992,
|
| 224 |
+
"grad_norm": 2.3588924407958984,
|
| 225 |
+
"learning_rate": 0.00022700327204347966,
|
| 226 |
+
"loss": 2.4169,
|
| 227 |
+
"step": 150
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.4116865869853918,
|
| 231 |
+
"grad_norm": 2.468594789505005,
|
| 232 |
+
"learning_rate": 0.00022093994792797152,
|
| 233 |
+
"loss": 2.3924,
|
| 234 |
+
"step": 155
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.4249667994687915,
|
| 238 |
+
"grad_norm": 2.52724552154541,
|
| 239 |
+
"learning_rate": 0.00021472343782058458,
|
| 240 |
+
"loss": 2.3804,
|
| 241 |
+
"step": 160
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.43824701195219123,
|
| 245 |
+
"grad_norm": 2.2549281120300293,
|
| 246 |
+
"learning_rate": 0.00020836716550149683,
|
| 247 |
+
"loss": 2.3681,
|
| 248 |
+
"step": 165
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.45152722443559096,
|
| 252 |
+
"grad_norm": 2.3330459594726562,
|
| 253 |
+
"learning_rate": 0.00020188485654999466,
|
| 254 |
+
"loss": 2.3404,
|
| 255 |
+
"step": 170
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.4648074369189907,
|
| 259 |
+
"grad_norm": 2.2739274501800537,
|
| 260 |
+
"learning_rate": 0.0001952905087057917,
|
| 261 |
+
"loss": 2.2986,
|
| 262 |
+
"step": 175
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.47808764940239046,
|
| 266 |
+
"grad_norm": 2.257761001586914,
|
| 267 |
+
"learning_rate": 0.00018859836164265164,
|
| 268 |
+
"loss": 2.328,
|
| 269 |
+
"step": 180
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.4913678618857902,
|
| 273 |
+
"grad_norm": 2.2927610874176025,
|
| 274 |
+
"learning_rate": 0.00018182286621958436,
|
| 275 |
+
"loss": 2.2641,
|
| 276 |
+
"step": 185
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.5046480743691899,
|
| 280 |
+
"grad_norm": 2.1379075050354004,
|
| 281 |
+
"learning_rate": 0.00017497865327601326,
|
| 282 |
+
"loss": 2.2607,
|
| 283 |
+
"step": 190
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.5179282868525896,
|
| 287 |
+
"grad_norm": 2.1380867958068848,
|
| 288 |
+
"learning_rate": 0.00016808050203829842,
|
| 289 |
+
"loss": 2.2726,
|
| 290 |
+
"step": 195
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.5312084993359893,
|
| 294 |
+
"grad_norm": 2.1610195636749268,
|
| 295 |
+
"learning_rate": 0.00016114330820583557,
|
| 296 |
+
"loss": 2.2329,
|
| 297 |
+
"step": 200
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.5444887118193891,
|
| 301 |
+
"grad_norm": 2.081423044204712,
|
| 302 |
+
"learning_rate": 0.00015418205178564621,
|
| 303 |
+
"loss": 2.2403,
|
| 304 |
+
"step": 205
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.5577689243027888,
|
| 308 |
+
"grad_norm": 2.0720622539520264,
|
| 309 |
+
"learning_rate": 0.0001472117647449168,
|
| 310 |
+
"loss": 2.2263,
|
| 311 |
+
"step": 210
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.5710491367861886,
|
| 315 |
+
"grad_norm": 2.028615713119507,
|
| 316 |
+
"learning_rate": 0.0001402474985513351,
|
| 317 |
+
"loss": 2.1766,
|
| 318 |
+
"step": 215
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.5843293492695883,
|
| 322 |
+
"grad_norm": 2.022357702255249,
|
| 323 |
+
"learning_rate": 0.00013330429167131863,
|
| 324 |
+
"loss": 2.1622,
|
| 325 |
+
"step": 220
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.5976095617529881,
|
| 329 |
+
"grad_norm": 1.9566619396209717,
|
| 330 |
+
"learning_rate": 0.00012639713709631706,
|
| 331 |
+
"loss": 2.141,
|
| 332 |
+
"step": 225
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.6108897742363878,
|
| 336 |
+
"grad_norm": 2.0382509231567383,
|
| 337 |
+
"learning_rate": 0.000119540949967313,
|
| 338 |
+
"loss": 2.1133,
|
| 339 |
+
"step": 230
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.6241699867197875,
|
| 343 |
+
"grad_norm": 2.006354570388794,
|
| 344 |
+
"learning_rate": 0.00011275053536743006,
|
| 345 |
+
"loss": 2.0986,
|
| 346 |
+
"step": 235
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.6374501992031872,
|
| 350 |
+
"grad_norm": 1.9956600666046143,
|
| 351 |
+
"learning_rate": 0.00010604055635219803,
|
| 352 |
+
"loss": 2.0761,
|
| 353 |
+
"step": 240
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.650730411686587,
|
| 357 |
+
"grad_norm": 1.9418437480926514,
|
| 358 |
+
"learning_rate": 9.942550228650876e-05,
|
| 359 |
+
"loss": 2.0824,
|
| 360 |
+
"step": 245
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.6640106241699867,
|
| 364 |
+
"grad_norm": 1.95418381690979,
|
| 365 |
+
"learning_rate": 9.291965755663522e-05,
|
| 366 |
+
"loss": 2.1084,
|
| 367 |
+
"step": 250
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.6772908366533864,
|
| 371 |
+
"grad_norm": 1.961089849472046,
|
| 372 |
+
"learning_rate": 8.653707072487629e-05,
|
| 373 |
+
"loss": 2.053,
|
| 374 |
+
"step": 255
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.6905710491367862,
|
| 378 |
+
"grad_norm": 1.9297434091567993,
|
| 379 |
+
"learning_rate": 8.02915241934347e-05,
|
| 380 |
+
"loss": 2.0788,
|
| 381 |
+
"step": 260
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.703851261620186,
|
| 385 |
+
"grad_norm": 1.9308726787567139,
|
| 386 |
+
"learning_rate": 7.419650444303376e-05,
|
| 387 |
+
"loss": 2.0137,
|
| 388 |
+
"step": 265
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.7171314741035857,
|
| 392 |
+
"grad_norm": 1.8884217739105225,
|
| 393 |
+
"learning_rate": 6.82651729105402e-05,
|
| 394 |
+
"loss": 2.025,
|
| 395 |
+
"step": 270
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.7304116865869854,
|
| 399 |
+
"grad_norm": 1.9265509843826294,
|
| 400 |
+
"learning_rate": 6.251033756847875e-05,
|
| 401 |
+
"loss": 1.9988,
|
| 402 |
+
"step": 275
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.7436918990703851,
|
| 406 |
+
"grad_norm": 1.8479654788970947,
|
| 407 |
+
"learning_rate": 5.694442526780891e-05,
|
| 408 |
+
"loss": 2.0355,
|
| 409 |
+
"step": 280
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.7569721115537849,
|
| 413 |
+
"grad_norm": 1.915022850036621,
|
| 414 |
+
"learning_rate": 5.157945490368621e-05,
|
| 415 |
+
"loss": 1.9948,
|
| 416 |
+
"step": 285
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.7702523240371846,
|
| 420 |
+
"grad_norm": 1.9624898433685303,
|
| 421 |
+
"learning_rate": 4.6427011462153554e-05,
|
| 422 |
+
"loss": 1.9742,
|
| 423 |
+
"step": 290
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.7835325365205843,
|
| 427 |
+
"grad_norm": 1.8428194522857666,
|
| 428 |
+
"learning_rate": 4.149822100380507e-05,
|
| 429 |
+
"loss": 1.9431,
|
| 430 |
+
"step": 295
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.796812749003984,
|
| 434 |
+
"grad_norm": 1.8078436851501465,
|
| 435 |
+
"learning_rate": 3.680372663844246e-05,
|
| 436 |
+
"loss": 1.9428,
|
| 437 |
+
"step": 300
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.8100929614873837,
|
| 441 |
+
"grad_norm": 1.8866031169891357,
|
| 442 |
+
"learning_rate": 3.235366554260351e-05,
|
| 443 |
+
"loss": 1.9581,
|
| 444 |
+
"step": 305
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.8233731739707836,
|
| 448 |
+
"grad_norm": 1.9002296924591064,
|
| 449 |
+
"learning_rate": 2.8157647069590534e-05,
|
| 450 |
+
"loss": 1.9669,
|
| 451 |
+
"step": 310
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.8366533864541833,
|
| 455 |
+
"grad_norm": 1.8117656707763672,
|
| 456 |
+
"learning_rate": 2.422473199926742e-05,
|
| 457 |
+
"loss": 1.9113,
|
| 458 |
+
"step": 315
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.849933598937583,
|
| 462 |
+
"grad_norm": 1.7977352142333984,
|
| 463 |
+
"learning_rate": 2.056341297243272e-05,
|
| 464 |
+
"loss": 1.9298,
|
| 465 |
+
"step": 320
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.8632138114209827,
|
| 469 |
+
"grad_norm": 1.824103593826294,
|
| 470 |
+
"learning_rate": 1.718159615201853e-05,
|
| 471 |
+
"loss": 1.8986,
|
| 472 |
+
"step": 325
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.8764940239043825,
|
| 476 |
+
"grad_norm": 1.8645001649856567,
|
| 477 |
+
"learning_rate": 1.4086584150715103e-05,
|
| 478 |
+
"loss": 1.9055,
|
| 479 |
+
"step": 330
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.8897742363877822,
|
| 483 |
+
"grad_norm": 1.8154819011688232,
|
| 484 |
+
"learning_rate": 1.1285060261887419e-05,
|
| 485 |
+
"loss": 1.9125,
|
| 486 |
+
"step": 335
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.9030544488711819,
|
| 490 |
+
"grad_norm": 1.8848274946212769,
|
| 491 |
+
"learning_rate": 8.783074027834514e-06,
|
| 492 |
+
"loss": 1.9312,
|
| 493 |
+
"step": 340
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.9163346613545816,
|
| 497 |
+
"grad_norm": 1.8445605039596558,
|
| 498 |
+
"learning_rate": 6.586028176555536e-06,
|
| 499 |
+
"loss": 1.9288,
|
| 500 |
+
"step": 345
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.9296148738379814,
|
| 504 |
+
"grad_norm": 1.8792498111724854,
|
| 505 |
+
"learning_rate": 4.698666955230862e-06,
|
| 506 |
+
"loss": 1.8988,
|
| 507 |
+
"step": 350
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.9428950863213812,
|
| 511 |
+
"grad_norm": 1.8407002687454224,
|
| 512 |
+
"learning_rate": 3.125065885610456e-06,
|
| 513 |
+
"loss": 1.8648,
|
| 514 |
+
"step": 355
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.9561752988047809,
|
| 518 |
+
"grad_norm": 1.7950658798217773,
|
| 519 |
+
"learning_rate": 1.8686229634317206e-06,
|
| 520 |
+
"loss": 1.9352,
|
| 521 |
+
"step": 360
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.9694555112881806,
|
| 525 |
+
"grad_norm": 1.8157298564910889,
|
| 526 |
+
"learning_rate": 9.320513208704638e-07,
|
| 527 |
+
"loss": 1.8703,
|
| 528 |
+
"step": 365
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.9827357237715804,
|
| 532 |
+
"grad_norm": 1.839713454246521,
|
| 533 |
+
"learning_rate": 3.173733678694834e-07,
|
| 534 |
+
"loss": 1.8878,
|
| 535 |
+
"step": 370
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.9960159362549801,
|
| 539 |
+
"grad_norm": 1.858101487159729,
|
| 540 |
+
"learning_rate": 2.5916424995919837e-08,
|
| 541 |
+
"loss": 1.9159,
|
| 542 |
+
"step": 375
|
| 543 |
+
}
|
| 544 |
+
],
|
| 545 |
+
"logging_steps": 5,
|
| 546 |
+
"max_steps": 376,
|
| 547 |
+
"num_input_tokens_seen": 0,
|
| 548 |
+
"num_train_epochs": 1,
|
| 549 |
+
"save_steps": 500,
|
| 550 |
+
"stateful_callbacks": {
|
| 551 |
+
"TrainerControl": {
|
| 552 |
+
"args": {
|
| 553 |
+
"should_epoch_stop": false,
|
| 554 |
+
"should_evaluate": false,
|
| 555 |
+
"should_log": false,
|
| 556 |
+
"should_save": true,
|
| 557 |
+
"should_training_stop": true
|
| 558 |
+
},
|
| 559 |
+
"attributes": {}
|
| 560 |
+
}
|
| 561 |
+
},
|
| 562 |
+
"total_flos": 4.656989204616577e+18,
|
| 563 |
+
"train_batch_size": 128,
|
| 564 |
+
"trial_name": null,
|
| 565 |
+
"trial_params": null
|
| 566 |
+
}
|