Upload checkpoint-500/trainer_state.json with huggingface_hub
Browse files
checkpoint-500/trainer_state.json
ADDED
|
@@ -0,0 +1,741 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.4938061535358429,
|
| 3 |
+
"best_model_checkpoint": "/home/ray/default/save/checkpoint-500",
|
| 4 |
+
"epoch": 0.43327556325823224,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.004332755632582322,
|
| 13 |
+
"grad_norm": 0.10313185304403305,
|
| 14 |
+
"learning_rate": 8.333333333333333e-07,
|
| 15 |
+
"loss": 1.0706,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.008665511265164644,
|
| 20 |
+
"grad_norm": 0.10810094326734543,
|
| 21 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 22 |
+
"loss": 1.0694,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.012998266897746967,
|
| 27 |
+
"grad_norm": 0.09731286019086838,
|
| 28 |
+
"learning_rate": 2.5e-06,
|
| 29 |
+
"loss": 1.0706,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01733102253032929,
|
| 34 |
+
"grad_norm": 0.11459868401288986,
|
| 35 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 36 |
+
"loss": 1.0772,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.021663778162911613,
|
| 41 |
+
"grad_norm": 0.10845116525888443,
|
| 42 |
+
"learning_rate": 4.166666666666667e-06,
|
| 43 |
+
"loss": 1.0809,
|
| 44 |
+
"step": 25
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.025996533795493933,
|
| 48 |
+
"grad_norm": 0.12091381102800369,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 1.0737,
|
| 51 |
+
"step": 30
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.030329289428076257,
|
| 55 |
+
"grad_norm": 0.12315661460161209,
|
| 56 |
+
"learning_rate": 5.833333333333334e-06,
|
| 57 |
+
"loss": 1.0675,
|
| 58 |
+
"step": 35
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.03466204506065858,
|
| 62 |
+
"grad_norm": 0.13054250180721283,
|
| 63 |
+
"learning_rate": 6.666666666666667e-06,
|
| 64 |
+
"loss": 1.0638,
|
| 65 |
+
"step": 40
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0389948006932409,
|
| 69 |
+
"grad_norm": 0.13414187729358673,
|
| 70 |
+
"learning_rate": 7.5e-06,
|
| 71 |
+
"loss": 1.0588,
|
| 72 |
+
"step": 45
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.043327556325823226,
|
| 76 |
+
"grad_norm": 0.12992985546588898,
|
| 77 |
+
"learning_rate": 8.333333333333334e-06,
|
| 78 |
+
"loss": 1.0512,
|
| 79 |
+
"step": 50
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.047660311958405546,
|
| 83 |
+
"grad_norm": 0.11500700563192368,
|
| 84 |
+
"learning_rate": 9.166666666666666e-06,
|
| 85 |
+
"loss": 1.0545,
|
| 86 |
+
"step": 55
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.05199306759098787,
|
| 90 |
+
"grad_norm": 0.10958714783191681,
|
| 91 |
+
"learning_rate": 1e-05,
|
| 92 |
+
"loss": 1.0357,
|
| 93 |
+
"step": 60
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.05632582322357019,
|
| 97 |
+
"grad_norm": 0.10853663086891174,
|
| 98 |
+
"learning_rate": 1.0833333333333334e-05,
|
| 99 |
+
"loss": 1.0268,
|
| 100 |
+
"step": 65
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.060658578856152515,
|
| 104 |
+
"grad_norm": 0.12498235702514648,
|
| 105 |
+
"learning_rate": 1.1666666666666668e-05,
|
| 106 |
+
"loss": 1.0135,
|
| 107 |
+
"step": 70
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.06499133448873484,
|
| 111 |
+
"grad_norm": 0.11716682463884354,
|
| 112 |
+
"learning_rate": 1.25e-05,
|
| 113 |
+
"loss": 1.0014,
|
| 114 |
+
"step": 75
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.06932409012131716,
|
| 118 |
+
"grad_norm": 0.09530466794967651,
|
| 119 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 120 |
+
"loss": 0.9824,
|
| 121 |
+
"step": 80
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.07365684575389948,
|
| 125 |
+
"grad_norm": 0.09298353642225266,
|
| 126 |
+
"learning_rate": 1.4166666666666668e-05,
|
| 127 |
+
"loss": 0.9676,
|
| 128 |
+
"step": 85
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.0779896013864818,
|
| 132 |
+
"grad_norm": 0.09433547407388687,
|
| 133 |
+
"learning_rate": 1.5e-05,
|
| 134 |
+
"loss": 0.9463,
|
| 135 |
+
"step": 90
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.08232235701906412,
|
| 139 |
+
"grad_norm": 0.09255563467741013,
|
| 140 |
+
"learning_rate": 1.5833333333333333e-05,
|
| 141 |
+
"loss": 0.9341,
|
| 142 |
+
"step": 95
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.08665511265164645,
|
| 146 |
+
"grad_norm": 0.09951213002204895,
|
| 147 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 148 |
+
"loss": 0.9209,
|
| 149 |
+
"step": 100
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.09098786828422877,
|
| 153 |
+
"grad_norm": 0.10951012372970581,
|
| 154 |
+
"learning_rate": 1.75e-05,
|
| 155 |
+
"loss": 0.8927,
|
| 156 |
+
"step": 105
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.09532062391681109,
|
| 160 |
+
"grad_norm": 0.110866479575634,
|
| 161 |
+
"learning_rate": 1.8333333333333333e-05,
|
| 162 |
+
"loss": 0.8755,
|
| 163 |
+
"step": 110
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.09965337954939342,
|
| 167 |
+
"grad_norm": 0.11648208647966385,
|
| 168 |
+
"learning_rate": 1.9166666666666667e-05,
|
| 169 |
+
"loss": 0.8453,
|
| 170 |
+
"step": 115
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.10398613518197573,
|
| 174 |
+
"grad_norm": 0.12016862630844116,
|
| 175 |
+
"learning_rate": 2e-05,
|
| 176 |
+
"loss": 0.8081,
|
| 177 |
+
"step": 120
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.10831889081455806,
|
| 181 |
+
"grad_norm": 0.12387488037347794,
|
| 182 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 183 |
+
"loss": 0.7784,
|
| 184 |
+
"step": 125
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.11265164644714037,
|
| 188 |
+
"grad_norm": 0.12779255211353302,
|
| 189 |
+
"learning_rate": 2.1666666666666667e-05,
|
| 190 |
+
"loss": 0.7353,
|
| 191 |
+
"step": 130
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.1169844020797227,
|
| 195 |
+
"grad_norm": 0.12649372220039368,
|
| 196 |
+
"learning_rate": 2.25e-05,
|
| 197 |
+
"loss": 0.7085,
|
| 198 |
+
"step": 135
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.12131715771230503,
|
| 202 |
+
"grad_norm": 0.1445430964231491,
|
| 203 |
+
"learning_rate": 2.3333333333333336e-05,
|
| 204 |
+
"loss": 0.6753,
|
| 205 |
+
"step": 140
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.12564991334488734,
|
| 209 |
+
"grad_norm": 0.1329505294561386,
|
| 210 |
+
"learning_rate": 2.4166666666666667e-05,
|
| 211 |
+
"loss": 0.6448,
|
| 212 |
+
"step": 145
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.12998266897746968,
|
| 216 |
+
"grad_norm": 0.13544394075870514,
|
| 217 |
+
"learning_rate": 2.5e-05,
|
| 218 |
+
"loss": 0.621,
|
| 219 |
+
"step": 150
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.134315424610052,
|
| 223 |
+
"grad_norm": 0.1351090520620346,
|
| 224 |
+
"learning_rate": 2.5833333333333336e-05,
|
| 225 |
+
"loss": 0.5997,
|
| 226 |
+
"step": 155
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.1386481802426343,
|
| 230 |
+
"grad_norm": 0.127303346991539,
|
| 231 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 232 |
+
"loss": 0.5791,
|
| 233 |
+
"step": 160
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.14298093587521662,
|
| 237 |
+
"grad_norm": 0.1476767510175705,
|
| 238 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 239 |
+
"loss": 0.57,
|
| 240 |
+
"step": 165
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.14731369150779897,
|
| 244 |
+
"grad_norm": 0.13856437802314758,
|
| 245 |
+
"learning_rate": 2.8333333333333335e-05,
|
| 246 |
+
"loss": 0.5645,
|
| 247 |
+
"step": 170
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.15164644714038128,
|
| 251 |
+
"grad_norm": 0.1533508449792862,
|
| 252 |
+
"learning_rate": 2.916666666666667e-05,
|
| 253 |
+
"loss": 0.5583,
|
| 254 |
+
"step": 175
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.1559792027729636,
|
| 258 |
+
"grad_norm": 0.13325001299381256,
|
| 259 |
+
"learning_rate": 3e-05,
|
| 260 |
+
"loss": 0.555,
|
| 261 |
+
"step": 180
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.16031195840554593,
|
| 265 |
+
"grad_norm": 0.13416974246501923,
|
| 266 |
+
"learning_rate": 3.0833333333333335e-05,
|
| 267 |
+
"loss": 0.5439,
|
| 268 |
+
"step": 185
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.16464471403812825,
|
| 272 |
+
"grad_norm": 0.1278882771730423,
|
| 273 |
+
"learning_rate": 3.1666666666666666e-05,
|
| 274 |
+
"loss": 0.537,
|
| 275 |
+
"step": 190
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.16897746967071056,
|
| 279 |
+
"grad_norm": 0.14047101140022278,
|
| 280 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 281 |
+
"loss": 0.5381,
|
| 282 |
+
"step": 195
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.1733102253032929,
|
| 286 |
+
"grad_norm": 0.13340455293655396,
|
| 287 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 288 |
+
"loss": 0.5344,
|
| 289 |
+
"step": 200
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.17764298093587522,
|
| 293 |
+
"grad_norm": 0.13049094378948212,
|
| 294 |
+
"learning_rate": 3.4166666666666666e-05,
|
| 295 |
+
"loss": 0.5291,
|
| 296 |
+
"step": 205
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.18197573656845753,
|
| 300 |
+
"grad_norm": 0.16296444833278656,
|
| 301 |
+
"learning_rate": 3.5e-05,
|
| 302 |
+
"loss": 0.5342,
|
| 303 |
+
"step": 210
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.18630849220103987,
|
| 307 |
+
"grad_norm": 0.1682613343000412,
|
| 308 |
+
"learning_rate": 3.5833333333333335e-05,
|
| 309 |
+
"loss": 0.5285,
|
| 310 |
+
"step": 215
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.19064124783362218,
|
| 314 |
+
"grad_norm": 0.1439386010169983,
|
| 315 |
+
"learning_rate": 3.6666666666666666e-05,
|
| 316 |
+
"loss": 0.5268,
|
| 317 |
+
"step": 220
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.1949740034662045,
|
| 321 |
+
"grad_norm": 0.15248768031597137,
|
| 322 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 323 |
+
"loss": 0.5252,
|
| 324 |
+
"step": 225
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.19930675909878684,
|
| 328 |
+
"grad_norm": 0.1604214906692505,
|
| 329 |
+
"learning_rate": 3.8333333333333334e-05,
|
| 330 |
+
"loss": 0.5201,
|
| 331 |
+
"step": 230
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.20363951473136915,
|
| 335 |
+
"grad_norm": 0.16192543506622314,
|
| 336 |
+
"learning_rate": 3.9166666666666665e-05,
|
| 337 |
+
"loss": 0.5222,
|
| 338 |
+
"step": 235
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.20797227036395147,
|
| 342 |
+
"grad_norm": 0.14945088326931,
|
| 343 |
+
"learning_rate": 4e-05,
|
| 344 |
+
"loss": 0.5158,
|
| 345 |
+
"step": 240
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.2123050259965338,
|
| 349 |
+
"grad_norm": 0.16204427182674408,
|
| 350 |
+
"learning_rate": 4.0833333333333334e-05,
|
| 351 |
+
"loss": 0.5127,
|
| 352 |
+
"step": 245
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.21663778162911612,
|
| 356 |
+
"grad_norm": 0.1618306040763855,
|
| 357 |
+
"learning_rate": 4.166666666666667e-05,
|
| 358 |
+
"loss": 0.5181,
|
| 359 |
+
"step": 250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.22097053726169844,
|
| 363 |
+
"grad_norm": 0.14193885028362274,
|
| 364 |
+
"learning_rate": 4.25e-05,
|
| 365 |
+
"loss": 0.5164,
|
| 366 |
+
"step": 255
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.22530329289428075,
|
| 370 |
+
"grad_norm": 0.12552691996097565,
|
| 371 |
+
"learning_rate": 4.3333333333333334e-05,
|
| 372 |
+
"loss": 0.5149,
|
| 373 |
+
"step": 260
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.2296360485268631,
|
| 377 |
+
"grad_norm": 0.14471225440502167,
|
| 378 |
+
"learning_rate": 4.4166666666666665e-05,
|
| 379 |
+
"loss": 0.5137,
|
| 380 |
+
"step": 265
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.2339688041594454,
|
| 384 |
+
"grad_norm": 0.13988590240478516,
|
| 385 |
+
"learning_rate": 4.5e-05,
|
| 386 |
+
"loss": 0.5066,
|
| 387 |
+
"step": 270
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.23830155979202772,
|
| 391 |
+
"grad_norm": 0.13964875042438507,
|
| 392 |
+
"learning_rate": 4.5833333333333334e-05,
|
| 393 |
+
"loss": 0.5116,
|
| 394 |
+
"step": 275
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.24263431542461006,
|
| 398 |
+
"grad_norm": 0.12847208976745605,
|
| 399 |
+
"learning_rate": 4.666666666666667e-05,
|
| 400 |
+
"loss": 0.5095,
|
| 401 |
+
"step": 280
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.24696707105719237,
|
| 405 |
+
"grad_norm": 0.13142219185829163,
|
| 406 |
+
"learning_rate": 4.75e-05,
|
| 407 |
+
"loss": 0.5077,
|
| 408 |
+
"step": 285
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.2512998266897747,
|
| 412 |
+
"grad_norm": 0.19098567962646484,
|
| 413 |
+
"learning_rate": 4.8333333333333334e-05,
|
| 414 |
+
"loss": 0.5099,
|
| 415 |
+
"step": 290
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.255632582322357,
|
| 419 |
+
"grad_norm": 0.1430283784866333,
|
| 420 |
+
"learning_rate": 4.9166666666666665e-05,
|
| 421 |
+
"loss": 0.5036,
|
| 422 |
+
"step": 295
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.25996533795493937,
|
| 426 |
+
"grad_norm": 0.12594453990459442,
|
| 427 |
+
"learning_rate": 5e-05,
|
| 428 |
+
"loss": 0.5085,
|
| 429 |
+
"step": 300
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.26429809358752165,
|
| 433 |
+
"grad_norm": 0.14377984404563904,
|
| 434 |
+
"learning_rate": 4.999577115486055e-05,
|
| 435 |
+
"loss": 0.4978,
|
| 436 |
+
"step": 305
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.268630849220104,
|
| 440 |
+
"grad_norm": 0.12468158453702927,
|
| 441 |
+
"learning_rate": 4.998308605009268e-05,
|
| 442 |
+
"loss": 0.5095,
|
| 443 |
+
"step": 310
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.2729636048526863,
|
| 447 |
+
"grad_norm": 0.12854167819023132,
|
| 448 |
+
"learning_rate": 4.996194897716389e-05,
|
| 449 |
+
"loss": 0.5037,
|
| 450 |
+
"step": 315
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.2772963604852686,
|
| 454 |
+
"grad_norm": 0.13529527187347412,
|
| 455 |
+
"learning_rate": 4.993236708690683e-05,
|
| 456 |
+
"loss": 0.5058,
|
| 457 |
+
"step": 320
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.28162911611785096,
|
| 461 |
+
"grad_norm": 0.1377994418144226,
|
| 462 |
+
"learning_rate": 4.9894350387100126e-05,
|
| 463 |
+
"loss": 0.4998,
|
| 464 |
+
"step": 325
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.28596187175043325,
|
| 468 |
+
"grad_norm": 0.14942322671413422,
|
| 469 |
+
"learning_rate": 4.984791173908267e-05,
|
| 470 |
+
"loss": 0.5007,
|
| 471 |
+
"step": 330
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2902946273830156,
|
| 475 |
+
"grad_norm": 0.1366725116968155,
|
| 476 |
+
"learning_rate": 4.9793066853402536e-05,
|
| 477 |
+
"loss": 0.5038,
|
| 478 |
+
"step": 335
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.29462738301559793,
|
| 482 |
+
"grad_norm": 0.13133087754249573,
|
| 483 |
+
"learning_rate": 4.9729834284501995e-05,
|
| 484 |
+
"loss": 0.5062,
|
| 485 |
+
"step": 340
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.2989601386481802,
|
| 489 |
+
"grad_norm": 0.11725670844316483,
|
| 490 |
+
"learning_rate": 4.965823542444037e-05,
|
| 491 |
+
"loss": 0.5025,
|
| 492 |
+
"step": 345
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.30329289428076256,
|
| 496 |
+
"grad_norm": 0.10628046840429306,
|
| 497 |
+
"learning_rate": 4.9578294495656965e-05,
|
| 498 |
+
"loss": 0.4999,
|
| 499 |
+
"step": 350
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.3076256499133449,
|
| 503 |
+
"grad_norm": 0.13826170563697815,
|
| 504 |
+
"learning_rate": 4.949003854277644e-05,
|
| 505 |
+
"loss": 0.4978,
|
| 506 |
+
"step": 355
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.3119584055459272,
|
| 510 |
+
"grad_norm": 0.1305851936340332,
|
| 511 |
+
"learning_rate": 4.9393497423459376e-05,
|
| 512 |
+
"loss": 0.4997,
|
| 513 |
+
"step": 360
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.31629116117850953,
|
| 517 |
+
"grad_norm": 0.11465763300657272,
|
| 518 |
+
"learning_rate": 4.928870379830124e-05,
|
| 519 |
+
"loss": 0.5037,
|
| 520 |
+
"step": 365
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.32062391681109187,
|
| 524 |
+
"grad_norm": 0.15975706279277802,
|
| 525 |
+
"learning_rate": 4.9175693119783013e-05,
|
| 526 |
+
"loss": 0.4982,
|
| 527 |
+
"step": 370
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.32495667244367415,
|
| 531 |
+
"grad_norm": 0.15360799431800842,
|
| 532 |
+
"learning_rate": 4.905450362027738e-05,
|
| 533 |
+
"loss": 0.5013,
|
| 534 |
+
"step": 375
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.3292894280762565,
|
| 538 |
+
"grad_norm": 0.14006198942661285,
|
| 539 |
+
"learning_rate": 4.8925176299114416e-05,
|
| 540 |
+
"loss": 0.5008,
|
| 541 |
+
"step": 380
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.33362218370883884,
|
| 545 |
+
"grad_norm": 0.14255651831626892,
|
| 546 |
+
"learning_rate": 4.878775490871121e-05,
|
| 547 |
+
"loss": 0.4975,
|
| 548 |
+
"step": 385
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.3379549393414211,
|
| 552 |
+
"grad_norm": 0.12012791633605957,
|
| 553 |
+
"learning_rate": 4.864228593977006e-05,
|
| 554 |
+
"loss": 0.5047,
|
| 555 |
+
"step": 390
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.34228769497400346,
|
| 559 |
+
"grad_norm": 0.11433300375938416,
|
| 560 |
+
"learning_rate": 4.848881860555035e-05,
|
| 561 |
+
"loss": 0.4986,
|
| 562 |
+
"step": 395
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.3466204506065858,
|
| 566 |
+
"grad_norm": 0.11102011054754257,
|
| 567 |
+
"learning_rate": 4.832740482521931e-05,
|
| 568 |
+
"loss": 0.4981,
|
| 569 |
+
"step": 400
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.3509532062391681,
|
| 573 |
+
"grad_norm": 0.12340573221445084,
|
| 574 |
+
"learning_rate": 4.815809920628738e-05,
|
| 575 |
+
"loss": 0.4984,
|
| 576 |
+
"step": 405
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.35528596187175043,
|
| 580 |
+
"grad_norm": 0.1148650050163269,
|
| 581 |
+
"learning_rate": 4.7980959026134044e-05,
|
| 582 |
+
"loss": 0.4942,
|
| 583 |
+
"step": 410
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3596187175043328,
|
| 587 |
+
"grad_norm": 0.10840712487697601,
|
| 588 |
+
"learning_rate": 4.7796044212630486e-05,
|
| 589 |
+
"loss": 0.4903,
|
| 590 |
+
"step": 415
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.36395147313691506,
|
| 594 |
+
"grad_norm": 0.11093516647815704,
|
| 595 |
+
"learning_rate": 4.7603417323865547e-05,
|
| 596 |
+
"loss": 0.4957,
|
| 597 |
+
"step": 420
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.3682842287694974,
|
| 601 |
+
"grad_norm": 0.1237047016620636,
|
| 602 |
+
"learning_rate": 4.74031435269818e-05,
|
| 603 |
+
"loss": 0.499,
|
| 604 |
+
"step": 425
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.37261698440207974,
|
| 608 |
+
"grad_norm": 0.11613244563341141,
|
| 609 |
+
"learning_rate": 4.7195290576129034e-05,
|
| 610 |
+
"loss": 0.4959,
|
| 611 |
+
"step": 430
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.37694974003466203,
|
| 615 |
+
"grad_norm": 0.10890854150056839,
|
| 616 |
+
"learning_rate": 4.697992878954255e-05,
|
| 617 |
+
"loss": 0.4944,
|
| 618 |
+
"step": 435
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.38128249566724437,
|
| 622 |
+
"grad_norm": 0.11364572495222092,
|
| 623 |
+
"learning_rate": 4.6757131025753886e-05,
|
| 624 |
+
"loss": 0.4909,
|
| 625 |
+
"step": 440
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.3856152512998267,
|
| 629 |
+
"grad_norm": 0.12619757652282715,
|
| 630 |
+
"learning_rate": 4.652697265894228e-05,
|
| 631 |
+
"loss": 0.4966,
|
| 632 |
+
"step": 445
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.389948006932409,
|
| 636 |
+
"grad_norm": 0.11669816076755524,
|
| 637 |
+
"learning_rate": 4.628953155343499e-05,
|
| 638 |
+
"loss": 0.4956,
|
| 639 |
+
"step": 450
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.39428076256499134,
|
| 643 |
+
"grad_norm": 0.13808482885360718,
|
| 644 |
+
"learning_rate": 4.604488803736523e-05,
|
| 645 |
+
"loss": 0.4973,
|
| 646 |
+
"step": 455
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.3986135181975737,
|
| 650 |
+
"grad_norm": 0.11171045899391174,
|
| 651 |
+
"learning_rate": 4.579312487549649e-05,
|
| 652 |
+
"loss": 0.4903,
|
| 653 |
+
"step": 460
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.40294627383015597,
|
| 657 |
+
"grad_norm": 0.11149395257234573,
|
| 658 |
+
"learning_rate": 4.553432724122265e-05,
|
| 659 |
+
"loss": 0.4999,
|
| 660 |
+
"step": 465
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.4072790294627383,
|
| 664 |
+
"grad_norm": 0.11308333277702332,
|
| 665 |
+
"learning_rate": 4.526858268775313e-05,
|
| 666 |
+
"loss": 0.4967,
|
| 667 |
+
"step": 470
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.41161178509532065,
|
| 671 |
+
"grad_norm": 0.10610105097293854,
|
| 672 |
+
"learning_rate": 4.499598111849299e-05,
|
| 673 |
+
"loss": 0.4936,
|
| 674 |
+
"step": 475
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.41594454072790293,
|
| 678 |
+
"grad_norm": 0.11356962472200394,
|
| 679 |
+
"learning_rate": 4.471661475662792e-05,
|
| 680 |
+
"loss": 0.493,
|
| 681 |
+
"step": 480
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.4202772963604853,
|
| 685 |
+
"grad_norm": 0.10310888290405273,
|
| 686 |
+
"learning_rate": 4.443057811392445e-05,
|
| 687 |
+
"loss": 0.5002,
|
| 688 |
+
"step": 485
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.4246100519930676,
|
| 692 |
+
"grad_norm": 0.11358631402254105,
|
| 693 |
+
"learning_rate": 4.413796795875586e-05,
|
| 694 |
+
"loss": 0.4983,
|
| 695 |
+
"step": 490
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.4289428076256499,
|
| 699 |
+
"grad_norm": 0.12575574219226837,
|
| 700 |
+
"learning_rate": 4.383888328336476e-05,
|
| 701 |
+
"loss": 0.4949,
|
| 702 |
+
"step": 495
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.43327556325823224,
|
| 706 |
+
"grad_norm": 0.09878399968147278,
|
| 707 |
+
"learning_rate": 4.3533425270373216e-05,
|
| 708 |
+
"loss": 0.4953,
|
| 709 |
+
"step": 500
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.43327556325823224,
|
| 713 |
+
"eval_loss": 0.4938061535358429,
|
| 714 |
+
"eval_runtime": 140.4236,
|
| 715 |
+
"eval_samples_per_second": 2.685,
|
| 716 |
+
"eval_steps_per_second": 0.677,
|
| 717 |
+
"step": 500
|
| 718 |
+
}
|
| 719 |
+
],
|
| 720 |
+
"logging_steps": 5,
|
| 721 |
+
"max_steps": 1154,
|
| 722 |
+
"num_input_tokens_seen": 0,
|
| 723 |
+
"num_train_epochs": 1,
|
| 724 |
+
"save_steps": 500,
|
| 725 |
+
"stateful_callbacks": {
|
| 726 |
+
"TrainerControl": {
|
| 727 |
+
"args": {
|
| 728 |
+
"should_epoch_stop": false,
|
| 729 |
+
"should_evaluate": false,
|
| 730 |
+
"should_log": false,
|
| 731 |
+
"should_save": true,
|
| 732 |
+
"should_training_stop": false
|
| 733 |
+
},
|
| 734 |
+
"attributes": {}
|
| 735 |
+
}
|
| 736 |
+
},
|
| 737 |
+
"total_flos": 3.49519346663424e+17,
|
| 738 |
+
"train_batch_size": 4,
|
| 739 |
+
"trial_name": null,
|
| 740 |
+
"trial_params": null
|
| 741 |
+
}
|