Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +672 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,672 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 90,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.022222222222222223,
|
| 13 |
+
"grad_norm": 0.04145513522429468,
|
| 14 |
+
"learning_rate": 5.555555555555556e-06,
|
| 15 |
+
"loss": 0.163,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.044444444444444446,
|
| 20 |
+
"grad_norm": 0.03806177997938909,
|
| 21 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 22 |
+
"loss": 0.1836,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.06666666666666667,
|
| 27 |
+
"grad_norm": 0.042251235266237294,
|
| 28 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 29 |
+
"loss": 0.1856,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.08888888888888889,
|
| 34 |
+
"grad_norm": 0.03855616390033849,
|
| 35 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 36 |
+
"loss": 0.1723,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1111111111111111,
|
| 41 |
+
"grad_norm": 0.03746547163614947,
|
| 42 |
+
"learning_rate": 2.777777777777778e-05,
|
| 43 |
+
"loss": 0.1807,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.13333333333333333,
|
| 48 |
+
"grad_norm": 0.04899948662680787,
|
| 49 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 50 |
+
"loss": 0.1749,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.15555555555555556,
|
| 55 |
+
"grad_norm": 0.043322070961254594,
|
| 56 |
+
"learning_rate": 3.888888888888889e-05,
|
| 57 |
+
"loss": 0.1793,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.17777777777777778,
|
| 62 |
+
"grad_norm": 0.05063967435578541,
|
| 63 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 64 |
+
"loss": 0.1584,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.2,
|
| 69 |
+
"grad_norm": 0.04466829068234921,
|
| 70 |
+
"learning_rate": 5e-05,
|
| 71 |
+
"loss": 0.15,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2222222222222222,
|
| 76 |
+
"grad_norm": 0.040258968404847786,
|
| 77 |
+
"learning_rate": 4.998119881260576e-05,
|
| 78 |
+
"loss": 0.1535,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.24444444444444444,
|
| 83 |
+
"grad_norm": 0.02894419013448755,
|
| 84 |
+
"learning_rate": 4.99248235291948e-05,
|
| 85 |
+
"loss": 0.1449,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.26666666666666666,
|
| 90 |
+
"grad_norm": 0.032138571629973876,
|
| 91 |
+
"learning_rate": 4.983095894354858e-05,
|
| 92 |
+
"loss": 0.1354,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.28888888888888886,
|
| 97 |
+
"grad_norm": 0.04078602334001387,
|
| 98 |
+
"learning_rate": 4.969974623692023e-05,
|
| 99 |
+
"loss": 0.1541,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.3111111111111111,
|
| 104 |
+
"grad_norm": 0.05032291478536776,
|
| 105 |
+
"learning_rate": 4.953138276568462e-05,
|
| 106 |
+
"loss": 0.1392,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.3333333333333333,
|
| 111 |
+
"grad_norm": 0.050090620555320695,
|
| 112 |
+
"learning_rate": 4.9326121764495596e-05,
|
| 113 |
+
"loss": 0.1444,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.35555555555555557,
|
| 118 |
+
"grad_norm": 0.035480633684579106,
|
| 119 |
+
"learning_rate": 4.9084271965397014e-05,
|
| 120 |
+
"loss": 0.1304,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.37777777777777777,
|
| 125 |
+
"grad_norm": 0.032726655928985764,
|
| 126 |
+
"learning_rate": 4.880619713346039e-05,
|
| 127 |
+
"loss": 0.1311,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.4,
|
| 132 |
+
"grad_norm": 0.029131540724124428,
|
| 133 |
+
"learning_rate": 4.849231551964771e-05,
|
| 134 |
+
"loss": 0.1223,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.4222222222222222,
|
| 139 |
+
"grad_norm": 0.02081489783305179,
|
| 140 |
+
"learning_rate": 4.814309923172227e-05,
|
| 141 |
+
"loss": 0.118,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.4444444444444444,
|
| 146 |
+
"grad_norm": 0.0210654978446037,
|
| 147 |
+
"learning_rate": 4.775907352415367e-05,
|
| 148 |
+
"loss": 0.1149,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.4666666666666667,
|
| 153 |
+
"grad_norm": 0.02408549187311914,
|
| 154 |
+
"learning_rate": 4.734081600808531e-05,
|
| 155 |
+
"loss": 0.1091,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.4888888888888889,
|
| 160 |
+
"grad_norm": 0.02376905383589948,
|
| 161 |
+
"learning_rate": 4.6888955782552274e-05,
|
| 162 |
+
"loss": 0.1028,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.5111111111111111,
|
| 167 |
+
"grad_norm": 0.02173656728014757,
|
| 168 |
+
"learning_rate": 4.640417248825667e-05,
|
| 169 |
+
"loss": 0.1119,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.5333333333333333,
|
| 174 |
+
"grad_norm": 0.02438990044693573,
|
| 175 |
+
"learning_rate": 4.588719528532342e-05,
|
| 176 |
+
"loss": 0.0992,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.5555555555555556,
|
| 181 |
+
"grad_norm": 0.022324375822038472,
|
| 182 |
+
"learning_rate": 4.533880175657419e-05,
|
| 183 |
+
"loss": 0.0978,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.5777777777777777,
|
| 188 |
+
"grad_norm": 0.021524779142548363,
|
| 189 |
+
"learning_rate": 4.475981673796899e-05,
|
| 190 |
+
"loss": 0.0941,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.6,
|
| 195 |
+
"grad_norm": 0.019110289084753145,
|
| 196 |
+
"learning_rate": 4.415111107797445e-05,
|
| 197 |
+
"loss": 0.0987,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.6222222222222222,
|
| 202 |
+
"grad_norm": 0.023272052760941467,
|
| 203 |
+
"learning_rate": 4.351360032772512e-05,
|
| 204 |
+
"loss": 0.0816,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.6444444444444445,
|
| 209 |
+
"grad_norm": 0.018818332739407733,
|
| 210 |
+
"learning_rate": 4.2848243363947484e-05,
|
| 211 |
+
"loss": 0.0877,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.6666666666666666,
|
| 216 |
+
"grad_norm": 0.022738471553081317,
|
| 217 |
+
"learning_rate": 4.215604094671835e-05,
|
| 218 |
+
"loss": 0.0729,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.6888888888888889,
|
| 223 |
+
"grad_norm": 0.016829351187921095,
|
| 224 |
+
"learning_rate": 4.14380342142266e-05,
|
| 225 |
+
"loss": 0.0793,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.7111111111111111,
|
| 230 |
+
"grad_norm": 0.021682647661793453,
|
| 231 |
+
"learning_rate": 4.069530311680247e-05,
|
| 232 |
+
"loss": 0.0926,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.7333333333333333,
|
| 237 |
+
"grad_norm": 0.020906839786747528,
|
| 238 |
+
"learning_rate": 3.9928964792569655e-05,
|
| 239 |
+
"loss": 0.0723,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.7555555555555555,
|
| 244 |
+
"grad_norm": 0.024600157953580535,
|
| 245 |
+
"learning_rate": 3.914017188716347e-05,
|
| 246 |
+
"loss": 0.0814,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.7777777777777778,
|
| 251 |
+
"grad_norm": 0.022668027885789463,
|
| 252 |
+
"learning_rate": 3.8330110820042285e-05,
|
| 253 |
+
"loss": 0.0866,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.8,
|
| 258 |
+
"grad_norm": 0.022565588311726368,
|
| 259 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 260 |
+
"loss": 0.086,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.8222222222222222,
|
| 265 |
+
"grad_norm": 0.01637375633073296,
|
| 266 |
+
"learning_rate": 3.665108799256348e-05,
|
| 267 |
+
"loss": 0.0745,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.8444444444444444,
|
| 272 |
+
"grad_norm": 0.016288962930045323,
|
| 273 |
+
"learning_rate": 3.578465164203134e-05,
|
| 274 |
+
"loss": 0.083,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.8666666666666667,
|
| 279 |
+
"grad_norm": 0.014072318472000743,
|
| 280 |
+
"learning_rate": 3.490199415097892e-05,
|
| 281 |
+
"loss": 0.0741,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.8888888888888888,
|
| 286 |
+
"grad_norm": 0.019458751327784993,
|
| 287 |
+
"learning_rate": 3.400444312011776e-05,
|
| 288 |
+
"loss": 0.071,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.9111111111111111,
|
| 293 |
+
"grad_norm": 0.016883319096903937,
|
| 294 |
+
"learning_rate": 3.309334855145803e-05,
|
| 295 |
+
"loss": 0.0751,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.9333333333333333,
|
| 300 |
+
"grad_norm": 0.01656670372950449,
|
| 301 |
+
"learning_rate": 3.217008081777726e-05,
|
| 302 |
+
"loss": 0.0736,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.9555555555555556,
|
| 307 |
+
"grad_norm": 0.01611572370876753,
|
| 308 |
+
"learning_rate": 3.1236028601449534e-05,
|
| 309 |
+
"loss": 0.0744,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.9777777777777777,
|
| 314 |
+
"grad_norm": 0.025130810955317162,
|
| 315 |
+
"learning_rate": 3.0292596805735274e-05,
|
| 316 |
+
"loss": 0.056,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.0,
|
| 321 |
+
"grad_norm": 0.018164615800081683,
|
| 322 |
+
"learning_rate": 2.9341204441673266e-05,
|
| 323 |
+
"loss": 0.0657,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.0222222222222221,
|
| 328 |
+
"grad_norm": 0.019962124083379607,
|
| 329 |
+
"learning_rate": 2.8383282493753283e-05,
|
| 330 |
+
"loss": 0.0567,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.0444444444444445,
|
| 335 |
+
"grad_norm": 0.016694411321077254,
|
| 336 |
+
"learning_rate": 2.742027176757948e-05,
|
| 337 |
+
"loss": 0.0656,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.0666666666666667,
|
| 342 |
+
"grad_norm": 0.0159764946766061,
|
| 343 |
+
"learning_rate": 2.6453620722761896e-05,
|
| 344 |
+
"loss": 0.0617,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.0888888888888888,
|
| 349 |
+
"grad_norm": 0.02018135193726075,
|
| 350 |
+
"learning_rate": 2.548478329429561e-05,
|
| 351 |
+
"loss": 0.0618,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.1111111111111112,
|
| 356 |
+
"grad_norm": 0.01398667566746435,
|
| 357 |
+
"learning_rate": 2.4515216705704395e-05,
|
| 358 |
+
"loss": 0.0659,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.1333333333333333,
|
| 363 |
+
"grad_norm": 0.015420358393515337,
|
| 364 |
+
"learning_rate": 2.3546379277238107e-05,
|
| 365 |
+
"loss": 0.0688,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.1555555555555554,
|
| 370 |
+
"grad_norm": 0.018423384035212806,
|
| 371 |
+
"learning_rate": 2.2579728232420525e-05,
|
| 372 |
+
"loss": 0.0683,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.1777777777777778,
|
| 377 |
+
"grad_norm": 0.014000938359142262,
|
| 378 |
+
"learning_rate": 2.161671750624673e-05,
|
| 379 |
+
"loss": 0.0575,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.2,
|
| 384 |
+
"grad_norm": 0.013842003203664193,
|
| 385 |
+
"learning_rate": 2.0658795558326743e-05,
|
| 386 |
+
"loss": 0.0617,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.2222222222222223,
|
| 391 |
+
"grad_norm": 0.02311016112104676,
|
| 392 |
+
"learning_rate": 1.970740319426474e-05,
|
| 393 |
+
"loss": 0.0711,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.2444444444444445,
|
| 398 |
+
"grad_norm": 0.01562457950516873,
|
| 399 |
+
"learning_rate": 1.876397139855047e-05,
|
| 400 |
+
"loss": 0.0615,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.2666666666666666,
|
| 405 |
+
"grad_norm": 0.02097714519013948,
|
| 406 |
+
"learning_rate": 1.7829919182222752e-05,
|
| 407 |
+
"loss": 0.0813,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.2888888888888888,
|
| 412 |
+
"grad_norm": 0.013954092588025305,
|
| 413 |
+
"learning_rate": 1.690665144854198e-05,
|
| 414 |
+
"loss": 0.0655,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.3111111111111111,
|
| 419 |
+
"grad_norm": 0.012050592569823772,
|
| 420 |
+
"learning_rate": 1.5995556879882246e-05,
|
| 421 |
+
"loss": 0.0687,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.3333333333333333,
|
| 426 |
+
"grad_norm": 0.012934161041963854,
|
| 427 |
+
"learning_rate": 1.509800584902108e-05,
|
| 428 |
+
"loss": 0.0604,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.3555555555555556,
|
| 433 |
+
"grad_norm": 0.01335553684986876,
|
| 434 |
+
"learning_rate": 1.4215348357968669e-05,
|
| 435 |
+
"loss": 0.0697,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.3777777777777778,
|
| 440 |
+
"grad_norm": 0.0128301661906966,
|
| 441 |
+
"learning_rate": 1.3348912007436537e-05,
|
| 442 |
+
"loss": 0.0661,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.4,
|
| 447 |
+
"grad_norm": 0.013663032199972917,
|
| 448 |
+
"learning_rate": 1.2500000000000006e-05,
|
| 449 |
+
"loss": 0.06,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.4222222222222223,
|
| 454 |
+
"grad_norm": 0.012134554930746227,
|
| 455 |
+
"learning_rate": 1.1669889179957725e-05,
|
| 456 |
+
"loss": 0.0582,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.4444444444444444,
|
| 461 |
+
"grad_norm": 0.014696339652119621,
|
| 462 |
+
"learning_rate": 1.085982811283654e-05,
|
| 463 |
+
"loss": 0.0536,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.4666666666666668,
|
| 468 |
+
"grad_norm": 0.01208908491764342,
|
| 469 |
+
"learning_rate": 1.0071035207430352e-05,
|
| 470 |
+
"loss": 0.0622,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.488888888888889,
|
| 475 |
+
"grad_norm": 0.013155964113476967,
|
| 476 |
+
"learning_rate": 9.304696883197542e-06,
|
| 477 |
+
"loss": 0.053,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.511111111111111,
|
| 482 |
+
"grad_norm": 0.010941706376727206,
|
| 483 |
+
"learning_rate": 8.561965785773413e-06,
|
| 484 |
+
"loss": 0.0611,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.5333333333333332,
|
| 489 |
+
"grad_norm": 0.011333878019293763,
|
| 490 |
+
"learning_rate": 7.843959053281663e-06,
|
| 491 |
+
"loss": 0.0708,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.5555555555555556,
|
| 496 |
+
"grad_norm": 0.012008866964762043,
|
| 497 |
+
"learning_rate": 7.1517566360525284e-06,
|
| 498 |
+
"loss": 0.0623,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.5777777777777777,
|
| 503 |
+
"grad_norm": 0.010450627576345212,
|
| 504 |
+
"learning_rate": 6.48639967227489e-06,
|
| 505 |
+
"loss": 0.0527,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.6,
|
| 510 |
+
"grad_norm": 0.01209040932177887,
|
| 511 |
+
"learning_rate": 5.848888922025553e-06,
|
| 512 |
+
"loss": 0.0731,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.6222222222222222,
|
| 517 |
+
"grad_norm": 0.016761186864218524,
|
| 518 |
+
"learning_rate": 5.240183262031021e-06,
|
| 519 |
+
"loss": 0.061,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.6444444444444444,
|
| 524 |
+
"grad_norm": 0.013223702181080193,
|
| 525 |
+
"learning_rate": 4.661198243425813e-06,
|
| 526 |
+
"loss": 0.0536,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.6666666666666665,
|
| 531 |
+
"grad_norm": 0.013217534982198317,
|
| 532 |
+
"learning_rate": 4.112804714676594e-06,
|
| 533 |
+
"loss": 0.0804,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.6888888888888889,
|
| 538 |
+
"grad_norm": 0.010298693910537537,
|
| 539 |
+
"learning_rate": 3.595827511743341e-06,
|
| 540 |
+
"loss": 0.0611,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.7111111111111112,
|
| 545 |
+
"grad_norm": 0.012020933219454272,
|
| 546 |
+
"learning_rate": 3.111044217447731e-06,
|
| 547 |
+
"loss": 0.0655,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.7333333333333334,
|
| 552 |
+
"grad_norm": 0.010506006734785647,
|
| 553 |
+
"learning_rate": 2.659183991914696e-06,
|
| 554 |
+
"loss": 0.0637,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.7555555555555555,
|
| 559 |
+
"grad_norm": 0.013713942520363884,
|
| 560 |
+
"learning_rate": 2.2409264758463363e-06,
|
| 561 |
+
"loss": 0.0692,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.7777777777777777,
|
| 566 |
+
"grad_norm": 0.010764341565662566,
|
| 567 |
+
"learning_rate": 1.8569007682777417e-06,
|
| 568 |
+
"loss": 0.0488,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.8,
|
| 573 |
+
"grad_norm": 0.01112279858087453,
|
| 574 |
+
"learning_rate": 1.5076844803522922e-06,
|
| 575 |
+
"loss": 0.053,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.8222222222222222,
|
| 580 |
+
"grad_norm": 0.011362400077819504,
|
| 581 |
+
"learning_rate": 1.1938028665396173e-06,
|
| 582 |
+
"loss": 0.0598,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.8444444444444446,
|
| 587 |
+
"grad_norm": 0.010591446954163518,
|
| 588 |
+
"learning_rate": 9.157280346029918e-07,
|
| 589 |
+
"loss": 0.0621,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.8666666666666667,
|
| 594 |
+
"grad_norm": 0.011204877929597866,
|
| 595 |
+
"learning_rate": 6.738782355044049e-07,
|
| 596 |
+
"loss": 0.0584,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.8888888888888888,
|
| 601 |
+
"grad_norm": 0.01257113405363327,
|
| 602 |
+
"learning_rate": 4.6861723431538276e-07,
|
| 603 |
+
"loss": 0.063,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.911111111111111,
|
| 608 |
+
"grad_norm": 0.01179877033535891,
|
| 609 |
+
"learning_rate": 3.002537630797747e-07,
|
| 610 |
+
"loss": 0.0561,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.9333333333333333,
|
| 615 |
+
"grad_norm": 0.011129626210418207,
|
| 616 |
+
"learning_rate": 1.6904105645142444e-07,
|
| 617 |
+
"loss": 0.0608,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.9555555555555557,
|
| 622 |
+
"grad_norm": 0.017223256380832427,
|
| 623 |
+
"learning_rate": 7.51764708051994e-08,
|
| 624 |
+
"loss": 0.0628,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.9777777777777779,
|
| 629 |
+
"grad_norm": 0.013357392038310047,
|
| 630 |
+
"learning_rate": 1.8801187394248965e-08,
|
| 631 |
+
"loss": 0.056,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.0,
|
| 636 |
+
"grad_norm": 0.014111952540712841,
|
| 637 |
+
"learning_rate": 0.0,
|
| 638 |
+
"loss": 0.0581,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.0,
|
| 643 |
+
"step": 90,
|
| 644 |
+
"total_flos": 1704342039035904.0,
|
| 645 |
+
"train_loss": 0.08834147482282585,
|
| 646 |
+
"train_runtime": 2715.1649,
|
| 647 |
+
"train_samples_per_second": 0.527,
|
| 648 |
+
"train_steps_per_second": 0.033
|
| 649 |
+
}
|
| 650 |
+
],
|
| 651 |
+
"logging_steps": 1,
|
| 652 |
+
"max_steps": 90,
|
| 653 |
+
"num_input_tokens_seen": 0,
|
| 654 |
+
"num_train_epochs": 2,
|
| 655 |
+
"save_steps": 500,
|
| 656 |
+
"stateful_callbacks": {
|
| 657 |
+
"TrainerControl": {
|
| 658 |
+
"args": {
|
| 659 |
+
"should_epoch_stop": false,
|
| 660 |
+
"should_evaluate": false,
|
| 661 |
+
"should_log": false,
|
| 662 |
+
"should_save": true,
|
| 663 |
+
"should_training_stop": true
|
| 664 |
+
},
|
| 665 |
+
"attributes": {}
|
| 666 |
+
}
|
| 667 |
+
},
|
| 668 |
+
"total_flos": 1704342039035904.0,
|
| 669 |
+
"train_batch_size": 2,
|
| 670 |
+
"trial_name": null,
|
| 671 |
+
"trial_params": null
|
| 672 |
+
}
|