generated from xuyuqing/ailab
2082 lines
48 KiB
JSON
2082 lines
48 KiB
JSON
{
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 181.8181818181818,
|
|
"global_step": 1000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 0.91,
|
|
"eval_loss": 2.799654722213745,
|
|
"eval_runtime": 0.6219,
|
|
"eval_samples_per_second": 283.023,
|
|
"eval_steps_per_second": 9.649,
|
|
"step": 5
|
|
},
|
|
{
|
|
"epoch": 1.82,
|
|
"learning_rate": 1.6666666666666667e-05,
|
|
"loss": 2.8028,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 2.0,
|
|
"eval_loss": 2.787478446960449,
|
|
"eval_runtime": 0.6249,
|
|
"eval_samples_per_second": 281.631,
|
|
"eval_steps_per_second": 9.601,
|
|
"step": 11
|
|
},
|
|
{
|
|
"epoch": 2.91,
|
|
"eval_loss": 2.7700254917144775,
|
|
"eval_runtime": 0.6268,
|
|
"eval_samples_per_second": 280.796,
|
|
"eval_steps_per_second": 9.573,
|
|
"step": 16
|
|
},
|
|
{
|
|
"epoch": 3.64,
|
|
"learning_rate": 3.3333333333333335e-05,
|
|
"loss": 2.7806,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 4.0,
|
|
"eval_loss": 2.7469289302825928,
|
|
"eval_runtime": 0.6233,
|
|
"eval_samples_per_second": 282.375,
|
|
"eval_steps_per_second": 9.626,
|
|
"step": 22
|
|
},
|
|
{
|
|
"epoch": 4.91,
|
|
"eval_loss": 2.7339789867401123,
|
|
"eval_runtime": 0.6397,
|
|
"eval_samples_per_second": 275.142,
|
|
"eval_steps_per_second": 9.38,
|
|
"step": 27
|
|
},
|
|
{
|
|
"epoch": 5.45,
|
|
"learning_rate": 5e-05,
|
|
"loss": 2.7377,
|
|
"step": 30
|
|
},
|
|
{
|
|
"epoch": 6.0,
|
|
"eval_loss": 2.718658685684204,
|
|
"eval_runtime": 0.6279,
|
|
"eval_samples_per_second": 280.283,
|
|
"eval_steps_per_second": 9.555,
|
|
"step": 33
|
|
},
|
|
{
|
|
"epoch": 6.91,
|
|
"eval_loss": 2.706568956375122,
|
|
"eval_runtime": 0.6452,
|
|
"eval_samples_per_second": 272.802,
|
|
"eval_steps_per_second": 9.3,
|
|
"step": 38
|
|
},
|
|
{
|
|
"epoch": 7.27,
|
|
"learning_rate": 4.998688922613788e-05,
|
|
"loss": 2.7233,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 8.0,
|
|
"eval_loss": 2.700272560119629,
|
|
"eval_runtime": 0.6918,
|
|
"eval_samples_per_second": 254.41,
|
|
"eval_steps_per_second": 8.673,
|
|
"step": 44
|
|
},
|
|
{
|
|
"epoch": 8.91,
|
|
"eval_loss": 2.6981542110443115,
|
|
"eval_runtime": 0.7009,
|
|
"eval_samples_per_second": 251.098,
|
|
"eval_steps_per_second": 8.56,
|
|
"step": 49
|
|
},
|
|
{
|
|
"epoch": 9.09,
|
|
"learning_rate": 4.9947570655942796e-05,
|
|
"loss": 2.7036,
|
|
"step": 50
|
|
},
|
|
{
|
|
"epoch": 10.0,
|
|
"eval_loss": 2.6956398487091064,
|
|
"eval_runtime": 0.6661,
|
|
"eval_samples_per_second": 264.224,
|
|
"eval_steps_per_second": 9.008,
|
|
"step": 55
|
|
},
|
|
{
|
|
"epoch": 10.91,
|
|
"learning_rate": 4.988208552916535e-05,
|
|
"loss": 2.6815,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 10.91,
|
|
"eval_loss": 2.690497398376465,
|
|
"eval_runtime": 0.6314,
|
|
"eval_samples_per_second": 278.763,
|
|
"eval_steps_per_second": 9.503,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 12.0,
|
|
"eval_loss": 2.6824758052825928,
|
|
"eval_runtime": 0.6907,
|
|
"eval_samples_per_second": 254.807,
|
|
"eval_steps_per_second": 8.687,
|
|
"step": 66
|
|
},
|
|
{
|
|
"epoch": 12.73,
|
|
"learning_rate": 4.9790502530660635e-05,
|
|
"loss": 2.6934,
|
|
"step": 70
|
|
},
|
|
{
|
|
"epoch": 12.91,
|
|
"eval_loss": 2.6752073764801025,
|
|
"eval_runtime": 0.7118,
|
|
"eval_samples_per_second": 247.267,
|
|
"eval_steps_per_second": 8.43,
|
|
"step": 71
|
|
},
|
|
{
|
|
"epoch": 14.0,
|
|
"eval_loss": 2.665703773498535,
|
|
"eval_runtime": 0.6671,
|
|
"eval_samples_per_second": 263.846,
|
|
"eval_steps_per_second": 8.995,
|
|
"step": 77
|
|
},
|
|
{
|
|
"epoch": 14.55,
|
|
"learning_rate": 4.967291771834727e-05,
|
|
"loss": 2.673,
|
|
"step": 80
|
|
},
|
|
{
|
|
"epoch": 14.91,
|
|
"eval_loss": 2.65787672996521,
|
|
"eval_runtime": 0.6312,
|
|
"eval_samples_per_second": 278.82,
|
|
"eval_steps_per_second": 9.505,
|
|
"step": 82
|
|
},
|
|
{
|
|
"epoch": 16.0,
|
|
"eval_loss": 2.6490914821624756,
|
|
"eval_runtime": 0.627,
|
|
"eval_samples_per_second": 280.719,
|
|
"eval_steps_per_second": 9.57,
|
|
"step": 88
|
|
},
|
|
{
|
|
"epoch": 16.36,
|
|
"learning_rate": 4.9529454422455976e-05,
|
|
"loss": 2.6601,
|
|
"step": 90
|
|
},
|
|
{
|
|
"epoch": 16.91,
|
|
"eval_loss": 2.6405282020568848,
|
|
"eval_runtime": 0.6302,
|
|
"eval_samples_per_second": 279.282,
|
|
"eval_steps_per_second": 9.521,
|
|
"step": 93
|
|
},
|
|
{
|
|
"epoch": 18.0,
|
|
"eval_loss": 2.6303794384002686,
|
|
"eval_runtime": 0.6268,
|
|
"eval_samples_per_second": 280.783,
|
|
"eval_steps_per_second": 9.572,
|
|
"step": 99
|
|
},
|
|
{
|
|
"epoch": 18.18,
|
|
"learning_rate": 4.936026311617316e-05,
|
|
"loss": 2.644,
|
|
"step": 100
|
|
},
|
|
{
|
|
"epoch": 18.91,
|
|
"eval_loss": 2.622063398361206,
|
|
"eval_runtime": 0.6388,
|
|
"eval_samples_per_second": 275.525,
|
|
"eval_steps_per_second": 9.393,
|
|
"step": 104
|
|
},
|
|
{
|
|
"epoch": 20.0,
|
|
"learning_rate": 4.916552125781528e-05,
|
|
"loss": 2.621,
|
|
"step": 110
|
|
},
|
|
{
|
|
"epoch": 20.0,
|
|
"eval_loss": 2.6123132705688477,
|
|
"eval_runtime": 0.6274,
|
|
"eval_samples_per_second": 280.519,
|
|
"eval_steps_per_second": 9.563,
|
|
"step": 110
|
|
},
|
|
{
|
|
"epoch": 20.91,
|
|
"eval_loss": 2.606389284133911,
|
|
"eval_runtime": 0.6277,
|
|
"eval_samples_per_second": 280.402,
|
|
"eval_steps_per_second": 9.559,
|
|
"step": 115
|
|
},
|
|
{
|
|
"epoch": 21.82,
|
|
"learning_rate": 4.894543310469968e-05,
|
|
"loss": 2.6036,
|
|
"step": 120
|
|
},
|
|
{
|
|
"epoch": 22.0,
|
|
"eval_loss": 2.5959715843200684,
|
|
"eval_runtime": 0.6275,
|
|
"eval_samples_per_second": 280.463,
|
|
"eval_steps_per_second": 9.561,
|
|
"step": 121
|
|
},
|
|
{
|
|
"epoch": 22.91,
|
|
"eval_loss": 2.5862438678741455,
|
|
"eval_runtime": 0.6276,
|
|
"eval_samples_per_second": 280.421,
|
|
"eval_steps_per_second": 9.56,
|
|
"step": 126
|
|
},
|
|
{
|
|
"epoch": 23.64,
|
|
"learning_rate": 4.870022949890676e-05,
|
|
"loss": 2.5607,
|
|
"step": 130
|
|
},
|
|
{
|
|
"epoch": 24.0,
|
|
"eval_loss": 2.5729258060455322,
|
|
"eval_runtime": 0.6262,
|
|
"eval_samples_per_second": 281.067,
|
|
"eval_steps_per_second": 9.582,
|
|
"step": 132
|
|
},
|
|
{
|
|
"epoch": 24.91,
|
|
"eval_loss": 2.5641725063323975,
|
|
"eval_runtime": 0.6306,
|
|
"eval_samples_per_second": 279.108,
|
|
"eval_steps_per_second": 9.515,
|
|
"step": 137
|
|
},
|
|
{
|
|
"epoch": 25.45,
|
|
"learning_rate": 4.8430167625158595e-05,
|
|
"loss": 2.5549,
|
|
"step": 140
|
|
},
|
|
{
|
|
"epoch": 26.0,
|
|
"eval_loss": 2.5501437187194824,
|
|
"eval_runtime": 0.6267,
|
|
"eval_samples_per_second": 280.837,
|
|
"eval_steps_per_second": 9.574,
|
|
"step": 143
|
|
},
|
|
{
|
|
"epoch": 26.91,
|
|
"eval_loss": 2.538557529449463,
|
|
"eval_runtime": 0.626,
|
|
"eval_samples_per_second": 281.146,
|
|
"eval_steps_per_second": 9.585,
|
|
"step": 148
|
|
},
|
|
{
|
|
"epoch": 27.27,
|
|
"learning_rate": 4.813553074106761e-05,
|
|
"loss": 2.5384,
|
|
"step": 150
|
|
},
|
|
{
|
|
"epoch": 28.0,
|
|
"eval_loss": 2.5269806385040283,
|
|
"eval_runtime": 0.6299,
|
|
"eval_samples_per_second": 279.409,
|
|
"eval_steps_per_second": 9.525,
|
|
"step": 154
|
|
},
|
|
{
|
|
"epoch": 28.91,
|
|
"eval_loss": 2.5143685340881348,
|
|
"eval_runtime": 0.6311,
|
|
"eval_samples_per_second": 278.859,
|
|
"eval_steps_per_second": 9.507,
|
|
"step": 159
|
|
},
|
|
{
|
|
"epoch": 29.09,
|
|
"learning_rate": 4.781662788003851e-05,
|
|
"loss": 2.5101,
|
|
"step": 160
|
|
},
|
|
{
|
|
"epoch": 30.0,
|
|
"eval_loss": 2.5024454593658447,
|
|
"eval_runtime": 0.625,
|
|
"eval_samples_per_second": 281.578,
|
|
"eval_steps_per_second": 9.599,
|
|
"step": 165
|
|
},
|
|
{
|
|
"epoch": 30.91,
|
|
"learning_rate": 4.747379352713489e-05,
|
|
"loss": 2.4797,
|
|
"step": 170
|
|
},
|
|
{
|
|
"epoch": 30.91,
|
|
"eval_loss": 2.490431308746338,
|
|
"eval_runtime": 0.6263,
|
|
"eval_samples_per_second": 281.011,
|
|
"eval_steps_per_second": 9.58,
|
|
"step": 170
|
|
},
|
|
{
|
|
"epoch": 32.0,
|
|
"eval_loss": 2.475470542907715,
|
|
"eval_runtime": 0.6391,
|
|
"eval_samples_per_second": 275.386,
|
|
"eval_steps_per_second": 9.388,
|
|
"step": 176
|
|
},
|
|
{
|
|
"epoch": 32.73,
|
|
"learning_rate": 4.710738726825059e-05,
|
|
"loss": 2.4571,
|
|
"step": 180
|
|
},
|
|
{
|
|
"epoch": 32.91,
|
|
"eval_loss": 2.4620444774627686,
|
|
"eval_runtime": 0.6276,
|
|
"eval_samples_per_second": 280.415,
|
|
"eval_steps_per_second": 9.56,
|
|
"step": 181
|
|
},
|
|
{
|
|
"epoch": 34.0,
|
|
"eval_loss": 2.4463887214660645,
|
|
"eval_runtime": 0.6276,
|
|
"eval_samples_per_second": 280.437,
|
|
"eval_steps_per_second": 9.56,
|
|
"step": 187
|
|
},
|
|
{
|
|
"epoch": 34.55,
|
|
"learning_rate": 4.671779341295378e-05,
|
|
"loss": 2.4163,
|
|
"step": 190
|
|
},
|
|
{
|
|
"epoch": 34.91,
|
|
"eval_loss": 2.4370341300964355,
|
|
"eval_runtime": 0.6288,
|
|
"eval_samples_per_second": 279.884,
|
|
"eval_steps_per_second": 9.541,
|
|
"step": 192
|
|
},
|
|
{
|
|
"epoch": 36.0,
|
|
"eval_loss": 2.4235072135925293,
|
|
"eval_runtime": 0.6315,
|
|
"eval_samples_per_second": 278.71,
|
|
"eval_steps_per_second": 9.501,
|
|
"step": 198
|
|
},
|
|
{
|
|
"epoch": 36.36,
|
|
"learning_rate": 4.630542059139924e-05,
|
|
"loss": 2.4053,
|
|
"step": 200
|
|
},
|
|
{
|
|
"epoch": 36.91,
|
|
"eval_loss": 2.4108152389526367,
|
|
"eval_runtime": 0.6286,
|
|
"eval_samples_per_second": 279.971,
|
|
"eval_steps_per_second": 9.544,
|
|
"step": 203
|
|
},
|
|
{
|
|
"epoch": 38.0,
|
|
"eval_loss": 2.395927667617798,
|
|
"eval_runtime": 0.6302,
|
|
"eval_samples_per_second": 279.26,
|
|
"eval_steps_per_second": 9.52,
|
|
"step": 209
|
|
},
|
|
{
|
|
"epoch": 38.18,
|
|
"learning_rate": 4.587070132573178e-05,
|
|
"loss": 2.3778,
|
|
"step": 210
|
|
},
|
|
{
|
|
"epoch": 38.91,
|
|
"eval_loss": 2.3836324214935303,
|
|
"eval_runtime": 0.6286,
|
|
"eval_samples_per_second": 279.984,
|
|
"eval_steps_per_second": 9.545,
|
|
"step": 214
|
|
},
|
|
{
|
|
"epoch": 40.0,
|
|
"learning_rate": 4.541409157643027e-05,
|
|
"loss": 2.3213,
|
|
"step": 220
|
|
},
|
|
{
|
|
"epoch": 40.0,
|
|
"eval_loss": 2.372097969055176,
|
|
"eval_runtime": 0.6279,
|
|
"eval_samples_per_second": 280.297,
|
|
"eval_steps_per_second": 9.556,
|
|
"step": 220
|
|
},
|
|
{
|
|
"epoch": 40.91,
|
|
"eval_loss": 2.3620593547821045,
|
|
"eval_runtime": 0.6282,
|
|
"eval_samples_per_second": 280.162,
|
|
"eval_steps_per_second": 9.551,
|
|
"step": 225
|
|
},
|
|
{
|
|
"epoch": 41.82,
|
|
"learning_rate": 4.493607026406802e-05,
|
|
"loss": 2.2901,
|
|
"step": 230
|
|
},
|
|
{
|
|
"epoch": 42.0,
|
|
"eval_loss": 2.350036859512329,
|
|
"eval_runtime": 0.6296,
|
|
"eval_samples_per_second": 279.526,
|
|
"eval_steps_per_second": 9.529,
|
|
"step": 231
|
|
},
|
|
{
|
|
"epoch": 42.91,
|
|
"eval_loss": 2.3378546237945557,
|
|
"eval_runtime": 0.6284,
|
|
"eval_samples_per_second": 280.057,
|
|
"eval_steps_per_second": 9.547,
|
|
"step": 236
|
|
},
|
|
{
|
|
"epoch": 43.64,
|
|
"learning_rate": 4.443713876699124e-05,
|
|
"loss": 2.2833,
|
|
"step": 240
|
|
},
|
|
{
|
|
"epoch": 44.0,
|
|
"eval_loss": 2.3216233253479004,
|
|
"eval_runtime": 0.6266,
|
|
"eval_samples_per_second": 280.88,
|
|
"eval_steps_per_second": 9.575,
|
|
"step": 242
|
|
},
|
|
{
|
|
"epoch": 44.91,
|
|
"eval_loss": 2.311659097671509,
|
|
"eval_runtime": 0.6303,
|
|
"eval_samples_per_second": 279.248,
|
|
"eval_steps_per_second": 9.52,
|
|
"step": 247
|
|
},
|
|
{
|
|
"epoch": 45.45,
|
|
"learning_rate": 4.391782039544238e-05,
|
|
"loss": 2.2398,
|
|
"step": 250
|
|
},
|
|
{
|
|
"epoch": 46.0,
|
|
"eval_loss": 2.292778491973877,
|
|
"eval_runtime": 0.6317,
|
|
"eval_samples_per_second": 278.625,
|
|
"eval_steps_per_second": 9.499,
|
|
"step": 253
|
|
},
|
|
{
|
|
"epoch": 46.91,
|
|
"eval_loss": 2.2805655002593994,
|
|
"eval_runtime": 0.6292,
|
|
"eval_samples_per_second": 279.712,
|
|
"eval_steps_per_second": 9.536,
|
|
"step": 258
|
|
},
|
|
{
|
|
"epoch": 47.27,
|
|
"learning_rate": 4.337865984268001e-05,
|
|
"loss": 2.2225,
|
|
"step": 260
|
|
},
|
|
{
|
|
"epoch": 48.0,
|
|
"eval_loss": 2.2669637203216553,
|
|
"eval_runtime": 0.6273,
|
|
"eval_samples_per_second": 280.565,
|
|
"eval_steps_per_second": 9.565,
|
|
"step": 264
|
|
},
|
|
{
|
|
"epoch": 48.91,
|
|
"eval_loss": 2.255608320236206,
|
|
"eval_runtime": 0.6357,
|
|
"eval_samples_per_second": 276.847,
|
|
"eval_steps_per_second": 9.438,
|
|
"step": 269
|
|
},
|
|
{
|
|
"epoch": 49.09,
|
|
"learning_rate": 4.2820222613670736e-05,
|
|
"loss": 2.1872,
|
|
"step": 270
|
|
},
|
|
{
|
|
"epoch": 50.0,
|
|
"eval_loss": 2.243396759033203,
|
|
"eval_runtime": 0.6308,
|
|
"eval_samples_per_second": 279.002,
|
|
"eval_steps_per_second": 9.511,
|
|
"step": 275
|
|
},
|
|
{
|
|
"epoch": 50.91,
|
|
"learning_rate": 4.224309443195261e-05,
|
|
"loss": 2.1735,
|
|
"step": 280
|
|
},
|
|
{
|
|
"epoch": 50.91,
|
|
"eval_loss": 2.233181953430176,
|
|
"eval_runtime": 0.628,
|
|
"eval_samples_per_second": 280.252,
|
|
"eval_steps_per_second": 9.554,
|
|
"step": 280
|
|
},
|
|
{
|
|
"epoch": 52.0,
|
|
"eval_loss": 2.224958896636963,
|
|
"eval_runtime": 0.6304,
|
|
"eval_samples_per_second": 279.173,
|
|
"eval_steps_per_second": 9.517,
|
|
"step": 286
|
|
},
|
|
{
|
|
"epoch": 52.73,
|
|
"learning_rate": 4.164788062529203e-05,
|
|
"loss": 2.1507,
|
|
"step": 290
|
|
},
|
|
{
|
|
"epoch": 52.91,
|
|
"eval_loss": 2.212779998779297,
|
|
"eval_runtime": 0.6276,
|
|
"eval_samples_per_second": 280.455,
|
|
"eval_steps_per_second": 9.561,
|
|
"step": 291
|
|
},
|
|
{
|
|
"epoch": 54.0,
|
|
"eval_loss": 2.2038145065307617,
|
|
"eval_runtime": 0.6283,
|
|
"eval_samples_per_second": 280.1,
|
|
"eval_steps_per_second": 9.549,
|
|
"step": 297
|
|
},
|
|
{
|
|
"epoch": 54.55,
|
|
"learning_rate": 4.10352054907785e-05,
|
|
"loss": 2.1199,
|
|
"step": 300
|
|
},
|
|
{
|
|
"epoch": 54.91,
|
|
"eval_loss": 2.1920149326324463,
|
|
"eval_runtime": 0.6274,
|
|
"eval_samples_per_second": 280.533,
|
|
"eval_steps_per_second": 9.564,
|
|
"step": 302
|
|
},
|
|
{
|
|
"epoch": 56.0,
|
|
"eval_loss": 2.1825506687164307,
|
|
"eval_runtime": 0.6362,
|
|
"eval_samples_per_second": 276.65,
|
|
"eval_steps_per_second": 9.431,
|
|
"step": 308
|
|
},
|
|
{
|
|
"epoch": 56.36,
|
|
"learning_rate": 4.0405711640023186e-05,
|
|
"loss": 2.0766,
|
|
"step": 310
|
|
},
|
|
{
|
|
"epoch": 56.91,
|
|
"eval_loss": 2.1739494800567627,
|
|
"eval_runtime": 0.6292,
|
|
"eval_samples_per_second": 279.702,
|
|
"eval_steps_per_second": 9.535,
|
|
"step": 313
|
|
},
|
|
{
|
|
"epoch": 58.0,
|
|
"eval_loss": 2.1604630947113037,
|
|
"eval_runtime": 0.6276,
|
|
"eval_samples_per_second": 280.418,
|
|
"eval_steps_per_second": 9.56,
|
|
"step": 319
|
|
},
|
|
{
|
|
"epoch": 58.18,
|
|
"learning_rate": 3.976005932514807e-05,
|
|
"loss": 2.0599,
|
|
"step": 320
|
|
},
|
|
{
|
|
"epoch": 58.91,
|
|
"eval_loss": 2.1482651233673096,
|
|
"eval_runtime": 0.6264,
|
|
"eval_samples_per_second": 280.98,
|
|
"eval_steps_per_second": 9.579,
|
|
"step": 324
|
|
},
|
|
{
|
|
"epoch": 60.0,
|
|
"learning_rate": 3.909892574627266e-05,
|
|
"loss": 2.0571,
|
|
"step": 330
|
|
},
|
|
{
|
|
"epoch": 60.0,
|
|
"eval_loss": 2.1340315341949463,
|
|
"eval_runtime": 0.6256,
|
|
"eval_samples_per_second": 281.335,
|
|
"eval_steps_per_second": 9.591,
|
|
"step": 330
|
|
},
|
|
{
|
|
"epoch": 60.91,
|
|
"eval_loss": 2.122722625732422,
|
|
"eval_runtime": 0.6262,
|
|
"eval_samples_per_second": 281.06,
|
|
"eval_steps_per_second": 9.582,
|
|
"step": 335
|
|
},
|
|
{
|
|
"epoch": 61.82,
|
|
"learning_rate": 3.84230043412246e-05,
|
|
"loss": 2.0039,
|
|
"step": 340
|
|
},
|
|
{
|
|
"epoch": 62.0,
|
|
"eval_loss": 2.1113879680633545,
|
|
"eval_runtime": 0.6353,
|
|
"eval_samples_per_second": 277.038,
|
|
"eval_steps_per_second": 9.444,
|
|
"step": 341
|
|
},
|
|
{
|
|
"epoch": 62.91,
|
|
"eval_loss": 2.103429079055786,
|
|
"eval_runtime": 0.6292,
|
|
"eval_samples_per_second": 279.718,
|
|
"eval_steps_per_second": 9.536,
|
|
"step": 346
|
|
},
|
|
{
|
|
"epoch": 63.64,
|
|
"learning_rate": 3.773300405821908e-05,
|
|
"loss": 1.9797,
|
|
"step": 350
|
|
},
|
|
{
|
|
"epoch": 64.0,
|
|
"eval_loss": 2.0910391807556152,
|
|
"eval_runtime": 0.6277,
|
|
"eval_samples_per_second": 280.376,
|
|
"eval_steps_per_second": 9.558,
|
|
"step": 352
|
|
},
|
|
{
|
|
"epoch": 64.91,
|
|
"eval_loss": 2.0817158222198486,
|
|
"eval_runtime": 0.6265,
|
|
"eval_samples_per_second": 280.945,
|
|
"eval_steps_per_second": 9.578,
|
|
"step": 357
|
|
},
|
|
{
|
|
"epoch": 65.45,
|
|
"learning_rate": 3.702964861227013e-05,
|
|
"loss": 1.9839,
|
|
"step": 360
|
|
},
|
|
{
|
|
"epoch": 66.0,
|
|
"eval_loss": 2.0717504024505615,
|
|
"eval_runtime": 0.6281,
|
|
"eval_samples_per_second": 280.224,
|
|
"eval_steps_per_second": 9.553,
|
|
"step": 363
|
|
},
|
|
{
|
|
"epoch": 66.91,
|
|
"eval_loss": 2.0628435611724854,
|
|
"eval_runtime": 0.6279,
|
|
"eval_samples_per_second": 280.279,
|
|
"eval_steps_per_second": 9.555,
|
|
"step": 368
|
|
},
|
|
{
|
|
"epoch": 67.27,
|
|
"learning_rate": 3.631367572611348e-05,
|
|
"loss": 1.9704,
|
|
"step": 370
|
|
},
|
|
{
|
|
"epoch": 68.0,
|
|
"eval_loss": 2.0526418685913086,
|
|
"eval_runtime": 0.6265,
|
|
"eval_samples_per_second": 280.947,
|
|
"eval_steps_per_second": 9.578,
|
|
"step": 374
|
|
},
|
|
{
|
|
"epoch": 68.91,
|
|
"eval_loss": 2.0455658435821533,
|
|
"eval_runtime": 0.6274,
|
|
"eval_samples_per_second": 280.513,
|
|
"eval_steps_per_second": 9.563,
|
|
"step": 379
|
|
},
|
|
{
|
|
"epoch": 69.09,
|
|
"learning_rate": 3.5585836356437264e-05,
|
|
"loss": 1.9156,
|
|
"step": 380
|
|
},
|
|
{
|
|
"epoch": 70.0,
|
|
"eval_loss": 2.034928798675537,
|
|
"eval_runtime": 0.7186,
|
|
"eval_samples_per_second": 244.921,
|
|
"eval_steps_per_second": 8.35,
|
|
"step": 385
|
|
},
|
|
{
|
|
"epoch": 70.91,
|
|
"learning_rate": 3.484689390623218e-05,
|
|
"loss": 1.9236,
|
|
"step": 390
|
|
},
|
|
{
|
|
"epoch": 70.91,
|
|
"eval_loss": 2.024853467941284,
|
|
"eval_runtime": 0.6793,
|
|
"eval_samples_per_second": 259.095,
|
|
"eval_steps_per_second": 8.833,
|
|
"step": 390
|
|
},
|
|
{
|
|
"epoch": 72.0,
|
|
"eval_loss": 2.014669179916382,
|
|
"eval_runtime": 0.6491,
|
|
"eval_samples_per_second": 271.151,
|
|
"eval_steps_per_second": 9.244,
|
|
"step": 396
|
|
},
|
|
{
|
|
"epoch": 72.73,
|
|
"learning_rate": 3.409762342408719e-05,
|
|
"loss": 1.9063,
|
|
"step": 400
|
|
},
|
|
{
|
|
"epoch": 72.91,
|
|
"eval_loss": 2.0030622482299805,
|
|
"eval_runtime": 0.6341,
|
|
"eval_samples_per_second": 277.58,
|
|
"eval_steps_per_second": 9.463,
|
|
"step": 401
|
|
},
|
|
{
|
|
"epoch": 74.0,
|
|
"eval_loss": 1.9951919317245483,
|
|
"eval_runtime": 0.629,
|
|
"eval_samples_per_second": 279.8,
|
|
"eval_steps_per_second": 9.539,
|
|
"step": 407
|
|
},
|
|
{
|
|
"epoch": 74.55,
|
|
"learning_rate": 3.333881079127052e-05,
|
|
"loss": 1.8796,
|
|
"step": 410
|
|
},
|
|
{
|
|
"epoch": 74.91,
|
|
"eval_loss": 1.987946629524231,
|
|
"eval_runtime": 0.6308,
|
|
"eval_samples_per_second": 279.016,
|
|
"eval_steps_per_second": 9.512,
|
|
"step": 412
|
|
},
|
|
{
|
|
"epoch": 76.0,
|
|
"eval_loss": 1.979250431060791,
|
|
"eval_runtime": 0.6296,
|
|
"eval_samples_per_second": 279.545,
|
|
"eval_steps_per_second": 9.53,
|
|
"step": 418
|
|
},
|
|
{
|
|
"epoch": 76.36,
|
|
"learning_rate": 3.2571251897448765e-05,
|
|
"loss": 1.8364,
|
|
"step": 420
|
|
},
|
|
{
|
|
"epoch": 76.91,
|
|
"eval_loss": 1.9724037647247314,
|
|
"eval_runtime": 0.636,
|
|
"eval_samples_per_second": 276.731,
|
|
"eval_steps_per_second": 9.434,
|
|
"step": 423
|
|
},
|
|
{
|
|
"epoch": 78.0,
|
|
"eval_loss": 1.9656604528427124,
|
|
"eval_runtime": 0.6304,
|
|
"eval_samples_per_second": 279.184,
|
|
"eval_steps_per_second": 9.518,
|
|
"step": 429
|
|
},
|
|
{
|
|
"epoch": 78.18,
|
|
"learning_rate": 3.1795751805908573e-05,
|
|
"loss": 1.8572,
|
|
"step": 430
|
|
},
|
|
{
|
|
"epoch": 78.91,
|
|
"eval_loss": 1.9555472135543823,
|
|
"eval_runtime": 0.6343,
|
|
"eval_samples_per_second": 277.464,
|
|
"eval_steps_per_second": 9.459,
|
|
"step": 434
|
|
},
|
|
{
|
|
"epoch": 80.0,
|
|
"learning_rate": 3.101312390915634e-05,
|
|
"loss": 1.8146,
|
|
"step": 440
|
|
},
|
|
{
|
|
"epoch": 80.0,
|
|
"eval_loss": 1.9422473907470703,
|
|
"eval_runtime": 0.6284,
|
|
"eval_samples_per_second": 280.08,
|
|
"eval_steps_per_second": 9.548,
|
|
"step": 440
|
|
},
|
|
{
|
|
"epoch": 80.91,
|
|
"eval_loss": 1.9326380491256714,
|
|
"eval_runtime": 0.6298,
|
|
"eval_samples_per_second": 279.472,
|
|
"eval_steps_per_second": 9.527,
|
|
"step": 445
|
|
},
|
|
{
|
|
"epoch": 81.82,
|
|
"learning_rate": 3.0224189075781884e-05,
|
|
"loss": 1.8065,
|
|
"step": 450
|
|
},
|
|
{
|
|
"epoch": 82.0,
|
|
"eval_loss": 1.9271011352539062,
|
|
"eval_runtime": 0.6302,
|
|
"eval_samples_per_second": 279.269,
|
|
"eval_steps_per_second": 9.521,
|
|
"step": 451
|
|
},
|
|
{
|
|
"epoch": 82.91,
|
|
"eval_loss": 1.9181721210479736,
|
|
"eval_runtime": 0.6289,
|
|
"eval_samples_per_second": 279.847,
|
|
"eval_steps_per_second": 9.54,
|
|
"step": 456
|
|
},
|
|
{
|
|
"epoch": 83.64,
|
|
"learning_rate": 2.9429774789480575e-05,
|
|
"loss": 1.7915,
|
|
"step": 460
|
|
},
|
|
{
|
|
"epoch": 84.0,
|
|
"eval_loss": 1.9119071960449219,
|
|
"eval_runtime": 0.6278,
|
|
"eval_samples_per_second": 280.36,
|
|
"eval_steps_per_second": 9.558,
|
|
"step": 462
|
|
},
|
|
{
|
|
"epoch": 84.91,
|
|
"eval_loss": 1.905703067779541,
|
|
"eval_runtime": 0.6294,
|
|
"eval_samples_per_second": 279.638,
|
|
"eval_steps_per_second": 9.533,
|
|
"step": 467
|
|
},
|
|
{
|
|
"epoch": 85.45,
|
|
"learning_rate": 2.863071428113726e-05,
|
|
"loss": 1.7727,
|
|
"step": 470
|
|
},
|
|
{
|
|
"epoch": 86.0,
|
|
"eval_loss": 1.8998371362686157,
|
|
"eval_runtime": 0.6307,
|
|
"eval_samples_per_second": 279.07,
|
|
"eval_steps_per_second": 9.514,
|
|
"step": 473
|
|
},
|
|
{
|
|
"epoch": 86.91,
|
|
"eval_loss": 1.8929120302200317,
|
|
"eval_runtime": 0.6362,
|
|
"eval_samples_per_second": 276.634,
|
|
"eval_steps_per_second": 9.431,
|
|
"step": 478
|
|
},
|
|
{
|
|
"epoch": 87.27,
|
|
"learning_rate": 2.782784565488211e-05,
|
|
"loss": 1.7504,
|
|
"step": 480
|
|
},
|
|
{
|
|
"epoch": 88.0,
|
|
"eval_loss": 1.888752818107605,
|
|
"eval_runtime": 0.6292,
|
|
"eval_samples_per_second": 279.7,
|
|
"eval_steps_per_second": 9.535,
|
|
"step": 484
|
|
},
|
|
{
|
|
"epoch": 88.91,
|
|
"eval_loss": 1.8814393281936646,
|
|
"eval_runtime": 0.6313,
|
|
"eval_samples_per_second": 278.792,
|
|
"eval_steps_per_second": 9.504,
|
|
"step": 489
|
|
},
|
|
{
|
|
"epoch": 89.09,
|
|
"learning_rate": 2.7022011009035107e-05,
|
|
"loss": 1.7473,
|
|
"step": 490
|
|
},
|
|
{
|
|
"epoch": 90.0,
|
|
"eval_loss": 1.874417781829834,
|
|
"eval_runtime": 0.6446,
|
|
"eval_samples_per_second": 273.047,
|
|
"eval_steps_per_second": 9.308,
|
|
"step": 495
|
|
},
|
|
{
|
|
"epoch": 90.91,
|
|
"learning_rate": 2.621405555286121e-05,
|
|
"loss": 1.7248,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 90.91,
|
|
"eval_loss": 1.8674060106277466,
|
|
"eval_runtime": 0.6218,
|
|
"eval_samples_per_second": 283.072,
|
|
"eval_steps_per_second": 9.65,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 92.0,
|
|
"eval_loss": 1.8550834655761719,
|
|
"eval_runtime": 0.6215,
|
|
"eval_samples_per_second": 283.203,
|
|
"eval_steps_per_second": 9.655,
|
|
"step": 506
|
|
},
|
|
{
|
|
"epoch": 92.73,
|
|
"learning_rate": 2.540482672006254e-05,
|
|
"loss": 1.7183,
|
|
"step": 510
|
|
},
|
|
{
|
|
"epoch": 92.91,
|
|
"eval_loss": 1.8496590852737427,
|
|
"eval_runtime": 0.6263,
|
|
"eval_samples_per_second": 281.011,
|
|
"eval_steps_per_second": 9.58,
|
|
"step": 511
|
|
},
|
|
{
|
|
"epoch": 94.0,
|
|
"eval_loss": 1.8428555727005005,
|
|
"eval_runtime": 0.6321,
|
|
"eval_samples_per_second": 278.449,
|
|
"eval_steps_per_second": 9.493,
|
|
"step": 517
|
|
},
|
|
{
|
|
"epoch": 94.55,
|
|
"learning_rate": 2.4595173279937464e-05,
|
|
"loss": 1.691,
|
|
"step": 520
|
|
},
|
|
{
|
|
"epoch": 94.91,
|
|
"eval_loss": 1.8376669883728027,
|
|
"eval_runtime": 0.6271,
|
|
"eval_samples_per_second": 280.653,
|
|
"eval_steps_per_second": 9.568,
|
|
"step": 522
|
|
},
|
|
{
|
|
"epoch": 96.0,
|
|
"eval_loss": 1.8278274536132812,
|
|
"eval_runtime": 0.6282,
|
|
"eval_samples_per_second": 280.171,
|
|
"eval_steps_per_second": 9.551,
|
|
"step": 528
|
|
},
|
|
{
|
|
"epoch": 96.36,
|
|
"learning_rate": 2.3785944447138802e-05,
|
|
"loss": 1.6836,
|
|
"step": 530
|
|
},
|
|
{
|
|
"epoch": 96.91,
|
|
"eval_loss": 1.8212929964065552,
|
|
"eval_runtime": 0.6282,
|
|
"eval_samples_per_second": 280.154,
|
|
"eval_steps_per_second": 9.551,
|
|
"step": 533
|
|
},
|
|
{
|
|
"epoch": 98.0,
|
|
"eval_loss": 1.813482642173767,
|
|
"eval_runtime": 0.6287,
|
|
"eval_samples_per_second": 279.935,
|
|
"eval_steps_per_second": 9.543,
|
|
"step": 539
|
|
},
|
|
{
|
|
"epoch": 98.18,
|
|
"learning_rate": 2.29779889909649e-05,
|
|
"loss": 1.7042,
|
|
"step": 540
|
|
},
|
|
{
|
|
"epoch": 98.91,
|
|
"eval_loss": 1.8071075677871704,
|
|
"eval_runtime": 0.6282,
|
|
"eval_samples_per_second": 280.145,
|
|
"eval_steps_per_second": 9.55,
|
|
"step": 544
|
|
},
|
|
{
|
|
"epoch": 100.0,
|
|
"learning_rate": 2.2172154345117894e-05,
|
|
"loss": 1.6415,
|
|
"step": 550
|
|
},
|
|
{
|
|
"epoch": 100.0,
|
|
"eval_loss": 1.8007694482803345,
|
|
"eval_runtime": 0.6278,
|
|
"eval_samples_per_second": 280.324,
|
|
"eval_steps_per_second": 9.557,
|
|
"step": 550
|
|
},
|
|
{
|
|
"epoch": 100.91,
|
|
"eval_loss": 1.796254277229309,
|
|
"eval_runtime": 0.6282,
|
|
"eval_samples_per_second": 280.155,
|
|
"eval_steps_per_second": 9.551,
|
|
"step": 555
|
|
},
|
|
{
|
|
"epoch": 101.82,
|
|
"learning_rate": 2.136928571886275e-05,
|
|
"loss": 1.6343,
|
|
"step": 560
|
|
},
|
|
{
|
|
"epoch": 102.0,
|
|
"eval_loss": 1.7895615100860596,
|
|
"eval_runtime": 0.6357,
|
|
"eval_samples_per_second": 276.869,
|
|
"eval_steps_per_second": 9.439,
|
|
"step": 561
|
|
},
|
|
{
|
|
"epoch": 102.91,
|
|
"eval_loss": 1.7855110168457031,
|
|
"eval_runtime": 0.6307,
|
|
"eval_samples_per_second": 279.077,
|
|
"eval_steps_per_second": 9.514,
|
|
"step": 566
|
|
},
|
|
{
|
|
"epoch": 103.64,
|
|
"learning_rate": 2.0570225210519434e-05,
|
|
"loss": 1.623,
|
|
"step": 570
|
|
},
|
|
{
|
|
"epoch": 104.0,
|
|
"eval_loss": 1.7835464477539062,
|
|
"eval_runtime": 0.63,
|
|
"eval_samples_per_second": 279.366,
|
|
"eval_steps_per_second": 9.524,
|
|
"step": 572
|
|
},
|
|
{
|
|
"epoch": 104.91,
|
|
"eval_loss": 1.7813255786895752,
|
|
"eval_runtime": 0.6414,
|
|
"eval_samples_per_second": 274.386,
|
|
"eval_steps_per_second": 9.354,
|
|
"step": 577
|
|
},
|
|
{
|
|
"epoch": 105.45,
|
|
"learning_rate": 1.9775810924218125e-05,
|
|
"loss": 1.6184,
|
|
"step": 580
|
|
},
|
|
{
|
|
"epoch": 106.0,
|
|
"eval_loss": 1.775755763053894,
|
|
"eval_runtime": 0.6424,
|
|
"eval_samples_per_second": 273.966,
|
|
"eval_steps_per_second": 9.34,
|
|
"step": 583
|
|
},
|
|
{
|
|
"epoch": 106.91,
|
|
"eval_loss": 1.7726260423660278,
|
|
"eval_runtime": 0.6294,
|
|
"eval_samples_per_second": 279.641,
|
|
"eval_steps_per_second": 9.533,
|
|
"step": 588
|
|
},
|
|
{
|
|
"epoch": 107.27,
|
|
"learning_rate": 1.8986876090843667e-05,
|
|
"loss": 1.6216,
|
|
"step": 590
|
|
},
|
|
{
|
|
"epoch": 108.0,
|
|
"eval_loss": 1.7669899463653564,
|
|
"eval_runtime": 0.6287,
|
|
"eval_samples_per_second": 279.942,
|
|
"eval_steps_per_second": 9.543,
|
|
"step": 594
|
|
},
|
|
{
|
|
"epoch": 108.91,
|
|
"eval_loss": 1.7590057849884033,
|
|
"eval_runtime": 0.6285,
|
|
"eval_samples_per_second": 280.031,
|
|
"eval_steps_per_second": 9.547,
|
|
"step": 599
|
|
},
|
|
{
|
|
"epoch": 109.09,
|
|
"learning_rate": 1.820424819409143e-05,
|
|
"loss": 1.5878,
|
|
"step": 600
|
|
},
|
|
{
|
|
"epoch": 110.0,
|
|
"eval_loss": 1.7512073516845703,
|
|
"eval_runtime": 0.6297,
|
|
"eval_samples_per_second": 279.506,
|
|
"eval_steps_per_second": 9.529,
|
|
"step": 605
|
|
},
|
|
{
|
|
"epoch": 110.91,
|
|
"learning_rate": 1.7428748102551237e-05,
|
|
"loss": 1.5909,
|
|
"step": 610
|
|
},
|
|
{
|
|
"epoch": 110.91,
|
|
"eval_loss": 1.7487860918045044,
|
|
"eval_runtime": 0.6281,
|
|
"eval_samples_per_second": 280.201,
|
|
"eval_steps_per_second": 9.552,
|
|
"step": 610
|
|
},
|
|
{
|
|
"epoch": 112.0,
|
|
"eval_loss": 1.744036078453064,
|
|
"eval_runtime": 0.6275,
|
|
"eval_samples_per_second": 280.464,
|
|
"eval_steps_per_second": 9.561,
|
|
"step": 616
|
|
},
|
|
{
|
|
"epoch": 112.73,
|
|
"learning_rate": 1.666118920872949e-05,
|
|
"loss": 1.5605,
|
|
"step": 620
|
|
},
|
|
{
|
|
"epoch": 112.91,
|
|
"eval_loss": 1.7392363548278809,
|
|
"eval_runtime": 0.6306,
|
|
"eval_samples_per_second": 279.117,
|
|
"eval_steps_per_second": 9.515,
|
|
"step": 621
|
|
},
|
|
{
|
|
"epoch": 114.0,
|
|
"eval_loss": 1.7366927862167358,
|
|
"eval_runtime": 0.6285,
|
|
"eval_samples_per_second": 280.01,
|
|
"eval_steps_per_second": 9.546,
|
|
"step": 627
|
|
},
|
|
{
|
|
"epoch": 114.55,
|
|
"learning_rate": 1.5902376575912815e-05,
|
|
"loss": 1.5619,
|
|
"step": 630
|
|
},
|
|
{
|
|
"epoch": 114.91,
|
|
"eval_loss": 1.733792781829834,
|
|
"eval_runtime": 0.6374,
|
|
"eval_samples_per_second": 276.124,
|
|
"eval_steps_per_second": 9.413,
|
|
"step": 632
|
|
},
|
|
{
|
|
"epoch": 116.0,
|
|
"eval_loss": 1.728451132774353,
|
|
"eval_runtime": 0.6285,
|
|
"eval_samples_per_second": 280.02,
|
|
"eval_steps_per_second": 9.546,
|
|
"step": 638
|
|
},
|
|
{
|
|
"epoch": 116.36,
|
|
"learning_rate": 1.5153106093767827e-05,
|
|
"loss": 1.5673,
|
|
"step": 640
|
|
},
|
|
{
|
|
"epoch": 116.91,
|
|
"eval_loss": 1.7250946760177612,
|
|
"eval_runtime": 0.6348,
|
|
"eval_samples_per_second": 277.248,
|
|
"eval_steps_per_second": 9.452,
|
|
"step": 643
|
|
},
|
|
{
|
|
"epoch": 118.0,
|
|
"eval_loss": 1.7209874391555786,
|
|
"eval_runtime": 0.695,
|
|
"eval_samples_per_second": 253.246,
|
|
"eval_steps_per_second": 8.633,
|
|
"step": 649
|
|
},
|
|
{
|
|
"epoch": 118.18,
|
|
"learning_rate": 1.4414163643562755e-05,
|
|
"loss": 1.5417,
|
|
"step": 650
|
|
},
|
|
{
|
|
"epoch": 118.91,
|
|
"eval_loss": 1.7177399396896362,
|
|
"eval_runtime": 0.7124,
|
|
"eval_samples_per_second": 247.048,
|
|
"eval_steps_per_second": 8.422,
|
|
"step": 654
|
|
},
|
|
{
|
|
"epoch": 120.0,
|
|
"learning_rate": 1.368632427388653e-05,
|
|
"loss": 1.5498,
|
|
"step": 660
|
|
},
|
|
{
|
|
"epoch": 120.0,
|
|
"eval_loss": 1.7115299701690674,
|
|
"eval_runtime": 0.6748,
|
|
"eval_samples_per_second": 260.823,
|
|
"eval_steps_per_second": 8.892,
|
|
"step": 660
|
|
},
|
|
{
|
|
"epoch": 120.91,
|
|
"eval_loss": 1.7086421251296997,
|
|
"eval_runtime": 0.6282,
|
|
"eval_samples_per_second": 280.181,
|
|
"eval_steps_per_second": 9.552,
|
|
"step": 665
|
|
},
|
|
{
|
|
"epoch": 121.82,
|
|
"learning_rate": 1.2970351387729873e-05,
|
|
"loss": 1.5398,
|
|
"step": 670
|
|
},
|
|
{
|
|
"epoch": 122.0,
|
|
"eval_loss": 1.702925443649292,
|
|
"eval_runtime": 0.6324,
|
|
"eval_samples_per_second": 278.318,
|
|
"eval_steps_per_second": 9.488,
|
|
"step": 671
|
|
},
|
|
{
|
|
"epoch": 122.91,
|
|
"eval_loss": 1.698160171508789,
|
|
"eval_runtime": 0.6286,
|
|
"eval_samples_per_second": 279.976,
|
|
"eval_steps_per_second": 9.545,
|
|
"step": 676
|
|
},
|
|
{
|
|
"epoch": 123.64,
|
|
"learning_rate": 1.2266995941780934e-05,
|
|
"loss": 1.532,
|
|
"step": 680
|
|
},
|
|
{
|
|
"epoch": 124.0,
|
|
"eval_loss": 1.694128155708313,
|
|
"eval_runtime": 0.6291,
|
|
"eval_samples_per_second": 279.744,
|
|
"eval_steps_per_second": 9.537,
|
|
"step": 682
|
|
},
|
|
{
|
|
"epoch": 124.91,
|
|
"eval_loss": 1.6900020837783813,
|
|
"eval_runtime": 0.6289,
|
|
"eval_samples_per_second": 279.861,
|
|
"eval_steps_per_second": 9.541,
|
|
"step": 687
|
|
},
|
|
{
|
|
"epoch": 125.45,
|
|
"learning_rate": 1.1576995658775405e-05,
|
|
"loss": 1.536,
|
|
"step": 690
|
|
},
|
|
{
|
|
"epoch": 126.0,
|
|
"eval_loss": 1.6851716041564941,
|
|
"eval_runtime": 0.6539,
|
|
"eval_samples_per_second": 269.135,
|
|
"eval_steps_per_second": 9.175,
|
|
"step": 693
|
|
},
|
|
{
|
|
"epoch": 126.91,
|
|
"eval_loss": 1.6837513446807861,
|
|
"eval_runtime": 0.6394,
|
|
"eval_samples_per_second": 275.259,
|
|
"eval_steps_per_second": 9.384,
|
|
"step": 698
|
|
},
|
|
{
|
|
"epoch": 127.27,
|
|
"learning_rate": 1.0901074253727336e-05,
|
|
"loss": 1.5134,
|
|
"step": 700
|
|
},
|
|
{
|
|
"epoch": 128.0,
|
|
"eval_loss": 1.6785069704055786,
|
|
"eval_runtime": 0.6314,
|
|
"eval_samples_per_second": 278.725,
|
|
"eval_steps_per_second": 9.502,
|
|
"step": 704
|
|
},
|
|
{
|
|
"epoch": 128.91,
|
|
"eval_loss": 1.6744980812072754,
|
|
"eval_runtime": 0.6297,
|
|
"eval_samples_per_second": 279.487,
|
|
"eval_steps_per_second": 9.528,
|
|
"step": 709
|
|
},
|
|
{
|
|
"epoch": 129.09,
|
|
"learning_rate": 1.0239940674851941e-05,
|
|
"loss": 1.515,
|
|
"step": 710
|
|
},
|
|
{
|
|
"epoch": 130.0,
|
|
"eval_loss": 1.6706637144088745,
|
|
"eval_runtime": 0.6291,
|
|
"eval_samples_per_second": 279.769,
|
|
"eval_steps_per_second": 9.538,
|
|
"step": 715
|
|
},
|
|
{
|
|
"epoch": 130.91,
|
|
"learning_rate": 9.594288359976817e-06,
|
|
"loss": 1.4918,
|
|
"step": 720
|
|
},
|
|
{
|
|
"epoch": 130.91,
|
|
"eval_loss": 1.6674672365188599,
|
|
"eval_runtime": 0.6288,
|
|
"eval_samples_per_second": 279.912,
|
|
"eval_steps_per_second": 9.542,
|
|
"step": 720
|
|
},
|
|
{
|
|
"epoch": 132.0,
|
|
"eval_loss": 1.6637736558914185,
|
|
"eval_runtime": 0.6306,
|
|
"eval_samples_per_second": 279.093,
|
|
"eval_steps_per_second": 9.515,
|
|
"step": 726
|
|
},
|
|
{
|
|
"epoch": 132.73,
|
|
"learning_rate": 8.964794509221508e-06,
|
|
"loss": 1.468,
|
|
"step": 730
|
|
},
|
|
{
|
|
"epoch": 132.91,
|
|
"eval_loss": 1.6616897583007812,
|
|
"eval_runtime": 0.6289,
|
|
"eval_samples_per_second": 279.838,
|
|
"eval_steps_per_second": 9.54,
|
|
"step": 731
|
|
},
|
|
{
|
|
"epoch": 134.0,
|
|
"eval_loss": 1.6599280834197998,
|
|
"eval_runtime": 0.6328,
|
|
"eval_samples_per_second": 278.119,
|
|
"eval_steps_per_second": 9.481,
|
|
"step": 737
|
|
},
|
|
{
|
|
"epoch": 134.55,
|
|
"learning_rate": 8.352119374707978e-06,
|
|
"loss": 1.4812,
|
|
"step": 740
|
|
},
|
|
{
|
|
"epoch": 134.91,
|
|
"eval_loss": 1.6583176851272583,
|
|
"eval_runtime": 0.6341,
|
|
"eval_samples_per_second": 277.571,
|
|
"eval_steps_per_second": 9.463,
|
|
"step": 742
|
|
},
|
|
{
|
|
"epoch": 136.0,
|
|
"eval_loss": 1.6539199352264404,
|
|
"eval_runtime": 0.629,
|
|
"eval_samples_per_second": 279.795,
|
|
"eval_steps_per_second": 9.538,
|
|
"step": 748
|
|
},
|
|
{
|
|
"epoch": 136.36,
|
|
"learning_rate": 7.756905568047393e-06,
|
|
"loss": 1.4931,
|
|
"step": 750
|
|
},
|
|
{
|
|
"epoch": 136.91,
|
|
"eval_loss": 1.650497317314148,
|
|
"eval_runtime": 0.63,
|
|
"eval_samples_per_second": 279.355,
|
|
"eval_steps_per_second": 9.523,
|
|
"step": 753
|
|
},
|
|
{
|
|
"epoch": 138.0,
|
|
"eval_loss": 1.6468048095703125,
|
|
"eval_runtime": 0.635,
|
|
"eval_samples_per_second": 277.184,
|
|
"eval_steps_per_second": 9.449,
|
|
"step": 759
|
|
},
|
|
{
|
|
"epoch": 138.18,
|
|
"learning_rate": 7.179777386329276e-06,
|
|
"loss": 1.4901,
|
|
"step": 760
|
|
},
|
|
{
|
|
"epoch": 138.91,
|
|
"eval_loss": 1.6437426805496216,
|
|
"eval_runtime": 0.6285,
|
|
"eval_samples_per_second": 280.025,
|
|
"eval_steps_per_second": 9.546,
|
|
"step": 764
|
|
},
|
|
{
|
|
"epoch": 140.0,
|
|
"learning_rate": 6.621340157319997e-06,
|
|
"loss": 1.4895,
|
|
"step": 770
|
|
},
|
|
{
|
|
"epoch": 140.0,
|
|
"eval_loss": 1.6408520936965942,
|
|
"eval_runtime": 0.6282,
|
|
"eval_samples_per_second": 280.144,
|
|
"eval_steps_per_second": 9.55,
|
|
"step": 770
|
|
},
|
|
{
|
|
"epoch": 140.91,
|
|
"eval_loss": 1.639256238937378,
|
|
"eval_runtime": 0.6291,
|
|
"eval_samples_per_second": 279.777,
|
|
"eval_steps_per_second": 9.538,
|
|
"step": 775
|
|
},
|
|
{
|
|
"epoch": 141.82,
|
|
"learning_rate": 6.082179604557617e-06,
|
|
"loss": 1.4622,
|
|
"step": 780
|
|
},
|
|
{
|
|
"epoch": 142.0,
|
|
"eval_loss": 1.6376827955245972,
|
|
"eval_runtime": 0.63,
|
|
"eval_samples_per_second": 279.371,
|
|
"eval_steps_per_second": 9.524,
|
|
"step": 781
|
|
},
|
|
{
|
|
"epoch": 142.91,
|
|
"eval_loss": 1.6356804370880127,
|
|
"eval_runtime": 0.6301,
|
|
"eval_samples_per_second": 279.341,
|
|
"eval_steps_per_second": 9.523,
|
|
"step": 786
|
|
},
|
|
{
|
|
"epoch": 143.64,
|
|
"learning_rate": 5.562861233008774e-06,
|
|
"loss": 1.4719,
|
|
"step": 790
|
|
},
|
|
{
|
|
"epoch": 144.0,
|
|
"eval_loss": 1.6334433555603027,
|
|
"eval_runtime": 0.6279,
|
|
"eval_samples_per_second": 280.303,
|
|
"eval_steps_per_second": 9.556,
|
|
"step": 792
|
|
},
|
|
{
|
|
"epoch": 144.91,
|
|
"eval_loss": 1.632299780845642,
|
|
"eval_runtime": 0.628,
|
|
"eval_samples_per_second": 280.244,
|
|
"eval_steps_per_second": 9.554,
|
|
"step": 797
|
|
},
|
|
{
|
|
"epoch": 145.45,
|
|
"learning_rate": 5.063929735931985e-06,
|
|
"loss": 1.46,
|
|
"step": 800
|
|
},
|
|
{
|
|
"epoch": 146.0,
|
|
"eval_loss": 1.6317532062530518,
|
|
"eval_runtime": 0.6313,
|
|
"eval_samples_per_second": 278.784,
|
|
"eval_steps_per_second": 9.504,
|
|
"step": 803
|
|
},
|
|
{
|
|
"epoch": 146.91,
|
|
"eval_loss": 1.6301318407058716,
|
|
"eval_runtime": 0.6356,
|
|
"eval_samples_per_second": 276.906,
|
|
"eval_steps_per_second": 9.44,
|
|
"step": 808
|
|
},
|
|
{
|
|
"epoch": 147.27,
|
|
"learning_rate": 4.585908423569724e-06,
|
|
"loss": 1.467,
|
|
"step": 810
|
|
},
|
|
{
|
|
"epoch": 148.0,
|
|
"eval_loss": 1.628135323524475,
|
|
"eval_runtime": 0.6281,
|
|
"eval_samples_per_second": 280.191,
|
|
"eval_steps_per_second": 9.552,
|
|
"step": 814
|
|
},
|
|
{
|
|
"epoch": 148.91,
|
|
"eval_loss": 1.627630352973938,
|
|
"eval_runtime": 0.6316,
|
|
"eval_samples_per_second": 278.65,
|
|
"eval_steps_per_second": 9.499,
|
|
"step": 819
|
|
},
|
|
{
|
|
"epoch": 149.09,
|
|
"learning_rate": 4.129298674268225e-06,
|
|
"loss": 1.4512,
|
|
"step": 820
|
|
},
|
|
{
|
|
"epoch": 150.0,
|
|
"eval_loss": 1.626612901687622,
|
|
"eval_runtime": 0.6301,
|
|
"eval_samples_per_second": 279.328,
|
|
"eval_steps_per_second": 9.523,
|
|
"step": 825
|
|
},
|
|
{
|
|
"epoch": 150.91,
|
|
"learning_rate": 3.694579408600771e-06,
|
|
"loss": 1.4578,
|
|
"step": 830
|
|
},
|
|
{
|
|
"epoch": 150.91,
|
|
"eval_loss": 1.6252334117889404,
|
|
"eval_runtime": 0.6286,
|
|
"eval_samples_per_second": 279.969,
|
|
"eval_steps_per_second": 9.544,
|
|
"step": 830
|
|
},
|
|
{
|
|
"epoch": 152.0,
|
|
"eval_loss": 1.623479962348938,
|
|
"eval_runtime": 0.6372,
|
|
"eval_samples_per_second": 276.212,
|
|
"eval_steps_per_second": 9.416,
|
|
"step": 836
|
|
},
|
|
{
|
|
"epoch": 152.73,
|
|
"learning_rate": 3.2822065870462217e-06,
|
|
"loss": 1.4448,
|
|
"step": 840
|
|
},
|
|
{
|
|
"epoch": 152.91,
|
|
"eval_loss": 1.6221522092819214,
|
|
"eval_runtime": 0.6304,
|
|
"eval_samples_per_second": 279.173,
|
|
"eval_steps_per_second": 9.517,
|
|
"step": 841
|
|
},
|
|
{
|
|
"epoch": 154.0,
|
|
"eval_loss": 1.621060848236084,
|
|
"eval_runtime": 0.6292,
|
|
"eval_samples_per_second": 279.72,
|
|
"eval_steps_per_second": 9.536,
|
|
"step": 847
|
|
},
|
|
{
|
|
"epoch": 154.55,
|
|
"learning_rate": 2.892612731749414e-06,
|
|
"loss": 1.4427,
|
|
"step": 850
|
|
},
|
|
{
|
|
"epoch": 154.91,
|
|
"eval_loss": 1.6197329759597778,
|
|
"eval_runtime": 0.6286,
|
|
"eval_samples_per_second": 279.976,
|
|
"eval_steps_per_second": 9.545,
|
|
"step": 852
|
|
},
|
|
{
|
|
"epoch": 156.0,
|
|
"eval_loss": 1.6186429262161255,
|
|
"eval_runtime": 0.6311,
|
|
"eval_samples_per_second": 278.886,
|
|
"eval_steps_per_second": 9.507,
|
|
"step": 858
|
|
},
|
|
{
|
|
"epoch": 156.36,
|
|
"learning_rate": 2.52620647286512e-06,
|
|
"loss": 1.4723,
|
|
"step": 860
|
|
},
|
|
{
|
|
"epoch": 156.91,
|
|
"eval_loss": 1.6177406311035156,
|
|
"eval_runtime": 0.6294,
|
|
"eval_samples_per_second": 279.628,
|
|
"eval_steps_per_second": 9.533,
|
|
"step": 863
|
|
},
|
|
{
|
|
"epoch": 158.0,
|
|
"eval_loss": 1.6171818971633911,
|
|
"eval_runtime": 0.628,
|
|
"eval_samples_per_second": 280.276,
|
|
"eval_steps_per_second": 9.555,
|
|
"step": 869
|
|
},
|
|
{
|
|
"epoch": 158.18,
|
|
"learning_rate": 2.183372119961499e-06,
|
|
"loss": 1.4621,
|
|
"step": 870
|
|
},
|
|
{
|
|
"epoch": 158.91,
|
|
"eval_loss": 1.6166049242019653,
|
|
"eval_runtime": 0.6298,
|
|
"eval_samples_per_second": 279.454,
|
|
"eval_steps_per_second": 9.527,
|
|
"step": 874
|
|
},
|
|
{
|
|
"epoch": 160.0,
|
|
"learning_rate": 1.864469258932397e-06,
|
|
"loss": 1.4333,
|
|
"step": 880
|
|
},
|
|
{
|
|
"epoch": 160.0,
|
|
"eval_loss": 1.6162300109863281,
|
|
"eval_runtime": 0.6291,
|
|
"eval_samples_per_second": 279.757,
|
|
"eval_steps_per_second": 9.537,
|
|
"step": 880
|
|
},
|
|
{
|
|
"epoch": 160.91,
|
|
"eval_loss": 1.61566162109375,
|
|
"eval_runtime": 0.6285,
|
|
"eval_samples_per_second": 280.027,
|
|
"eval_steps_per_second": 9.546,
|
|
"step": 885
|
|
},
|
|
{
|
|
"epoch": 161.82,
|
|
"learning_rate": 1.5698323748414124e-06,
|
|
"loss": 1.4486,
|
|
"step": 890
|
|
},
|
|
{
|
|
"epoch": 162.0,
|
|
"eval_loss": 1.6152541637420654,
|
|
"eval_runtime": 0.6273,
|
|
"eval_samples_per_second": 280.568,
|
|
"eval_steps_per_second": 9.565,
|
|
"step": 891
|
|
},
|
|
{
|
|
"epoch": 162.91,
|
|
"eval_loss": 1.614918828010559,
|
|
"eval_runtime": 0.6283,
|
|
"eval_samples_per_second": 280.126,
|
|
"eval_steps_per_second": 9.55,
|
|
"step": 896
|
|
},
|
|
{
|
|
"epoch": 163.64,
|
|
"learning_rate": 1.2997705010932393e-06,
|
|
"loss": 1.442,
|
|
"step": 900
|
|
},
|
|
{
|
|
"epoch": 164.0,
|
|
"eval_loss": 1.6146750450134277,
|
|
"eval_runtime": 0.6288,
|
|
"eval_samples_per_second": 279.907,
|
|
"eval_steps_per_second": 9.542,
|
|
"step": 902
|
|
},
|
|
{
|
|
"epoch": 164.91,
|
|
"eval_loss": 1.6144887208938599,
|
|
"eval_runtime": 0.6347,
|
|
"eval_samples_per_second": 277.317,
|
|
"eval_steps_per_second": 9.454,
|
|
"step": 907
|
|
},
|
|
{
|
|
"epoch": 165.45,
|
|
"learning_rate": 1.0545668953003241e-06,
|
|
"loss": 1.4536,
|
|
"step": 910
|
|
},
|
|
{
|
|
"epoch": 166.0,
|
|
"eval_loss": 1.6138886213302612,
|
|
"eval_runtime": 0.6294,
|
|
"eval_samples_per_second": 279.623,
|
|
"eval_steps_per_second": 9.533,
|
|
"step": 913
|
|
},
|
|
{
|
|
"epoch": 166.91,
|
|
"eval_loss": 1.6135444641113281,
|
|
"eval_runtime": 0.6307,
|
|
"eval_samples_per_second": 279.05,
|
|
"eval_steps_per_second": 9.513,
|
|
"step": 918
|
|
},
|
|
{
|
|
"epoch": 167.27,
|
|
"learning_rate": 8.344787421847217e-07,
|
|
"loss": 1.4393,
|
|
"step": 920
|
|
},
|
|
{
|
|
"epoch": 168.0,
|
|
"eval_loss": 1.6134400367736816,
|
|
"eval_runtime": 0.6358,
|
|
"eval_samples_per_second": 276.8,
|
|
"eval_steps_per_second": 9.436,
|
|
"step": 924
|
|
},
|
|
{
|
|
"epoch": 168.91,
|
|
"eval_loss": 1.6133301258087158,
|
|
"eval_runtime": 0.6303,
|
|
"eval_samples_per_second": 279.234,
|
|
"eval_steps_per_second": 9.519,
|
|
"step": 929
|
|
},
|
|
{
|
|
"epoch": 169.09,
|
|
"learning_rate": 6.397368838268497e-07,
|
|
"loss": 1.4096,
|
|
"step": 930
|
|
},
|
|
{
|
|
"epoch": 170.0,
|
|
"eval_loss": 1.6129955053329468,
|
|
"eval_runtime": 0.6409,
|
|
"eval_samples_per_second": 274.593,
|
|
"eval_steps_per_second": 9.361,
|
|
"step": 935
|
|
},
|
|
{
|
|
"epoch": 170.91,
|
|
"learning_rate": 4.7054557754402373e-07,
|
|
"loss": 1.4455,
|
|
"step": 940
|
|
},
|
|
{
|
|
"epoch": 170.91,
|
|
"eval_loss": 1.6126748323440552,
|
|
"eval_runtime": 0.6322,
|
|
"eval_samples_per_second": 278.399,
|
|
"eval_steps_per_second": 9.491,
|
|
"step": 940
|
|
},
|
|
{
|
|
"epoch": 172.0,
|
|
"eval_loss": 1.6124964952468872,
|
|
"eval_runtime": 0.6278,
|
|
"eval_samples_per_second": 280.359,
|
|
"eval_steps_per_second": 9.558,
|
|
"step": 946
|
|
},
|
|
{
|
|
"epoch": 172.73,
|
|
"learning_rate": 3.270822816527325e-07,
|
|
"loss": 1.4454,
|
|
"step": 950
|
|
},
|
|
{
|
|
"epoch": 172.91,
|
|
"eval_loss": 1.612331748008728,
|
|
"eval_runtime": 0.6344,
|
|
"eval_samples_per_second": 277.444,
|
|
"eval_steps_per_second": 9.458,
|
|
"step": 951
|
|
},
|
|
{
|
|
"epoch": 174.0,
|
|
"eval_loss": 1.612149715423584,
|
|
"eval_runtime": 0.6297,
|
|
"eval_samples_per_second": 279.494,
|
|
"eval_steps_per_second": 9.528,
|
|
"step": 957
|
|
},
|
|
{
|
|
"epoch": 174.55,
|
|
"learning_rate": 2.094974693393731e-07,
|
|
"loss": 1.4622,
|
|
"step": 960
|
|
},
|
|
{
|
|
"epoch": 174.91,
|
|
"eval_loss": 1.6121467351913452,
|
|
"eval_runtime": 0.7127,
|
|
"eval_samples_per_second": 246.947,
|
|
"eval_steps_per_second": 8.419,
|
|
"step": 962
|
|
},
|
|
{
|
|
"epoch": 176.0,
|
|
"eval_loss": 1.6120716333389282,
|
|
"eval_runtime": 0.7177,
|
|
"eval_samples_per_second": 245.216,
|
|
"eval_steps_per_second": 8.36,
|
|
"step": 968
|
|
},
|
|
{
|
|
"epoch": 176.36,
|
|
"learning_rate": 1.1791447083465134e-07,
|
|
"loss": 1.4356,
|
|
"step": 970
|
|
},
|
|
{
|
|
"epoch": 176.91,
|
|
"eval_loss": 1.6119561195373535,
|
|
"eval_runtime": 0.7008,
|
|
"eval_samples_per_second": 251.146,
|
|
"eval_steps_per_second": 8.562,
|
|
"step": 973
|
|
},
|
|
{
|
|
"epoch": 178.0,
|
|
"eval_loss": 1.6119166612625122,
|
|
"eval_runtime": 0.6284,
|
|
"eval_samples_per_second": 280.08,
|
|
"eval_steps_per_second": 9.548,
|
|
"step": 979
|
|
},
|
|
{
|
|
"epoch": 178.18,
|
|
"learning_rate": 5.242934405720879e-08,
|
|
"loss": 1.4389,
|
|
"step": 980
|
|
},
|
|
{
|
|
"epoch": 178.91,
|
|
"eval_loss": 1.6118892431259155,
|
|
"eval_runtime": 0.6324,
|
|
"eval_samples_per_second": 278.318,
|
|
"eval_steps_per_second": 9.488,
|
|
"step": 984
|
|
},
|
|
{
|
|
"epoch": 180.0,
|
|
"learning_rate": 1.3110773862126669e-08,
|
|
"loss": 1.4313,
|
|
"step": 990
|
|
},
|
|
{
|
|
"epoch": 180.0,
|
|
"eval_loss": 1.6118674278259277,
|
|
"eval_runtime": 0.6325,
|
|
"eval_samples_per_second": 278.263,
|
|
"eval_steps_per_second": 9.486,
|
|
"step": 990
|
|
},
|
|
{
|
|
"epoch": 180.91,
|
|
"eval_loss": 1.6118358373641968,
|
|
"eval_runtime": 0.6302,
|
|
"eval_samples_per_second": 279.257,
|
|
"eval_steps_per_second": 9.52,
|
|
"step": 995
|
|
},
|
|
{
|
|
"epoch": 181.82,
|
|
"learning_rate": 0.0,
|
|
"loss": 1.452,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 181.82,
|
|
"eval_loss": 1.6118431091308594,
|
|
"eval_runtime": 0.6469,
|
|
"eval_samples_per_second": 272.078,
|
|
"eval_steps_per_second": 9.275,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 181.82,
|
|
"step": 1000,
|
|
"total_flos": 1.303052976570499e+18,
|
|
"train_loss": 1.8824593620300294,
|
|
"train_runtime": 1161.1003,
|
|
"train_samples_per_second": 121.092,
|
|
"train_steps_per_second": 0.861
|
|
}
|
|
],
|
|
"max_steps": 1000,
|
|
"num_train_epochs": 200,
|
|
"total_flos": 1.303052976570499e+18,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|