generated from xuyuqing/ailab
2106 lines
50 KiB
JSON
2106 lines
50 KiB
JSON
{
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 200.0,
|
|
"global_step": 800,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 1.0,
|
|
"eval_loss": 2.7863576412200928,
|
|
"eval_runtime": 0.2338,
|
|
"eval_samples_per_second": 530.381,
|
|
"eval_steps_per_second": 17.109,
|
|
"step": 4
|
|
},
|
|
{
|
|
"epoch": 2.0,
|
|
"eval_loss": 2.7510902881622314,
|
|
"eval_runtime": 0.2316,
|
|
"eval_samples_per_second": 535.486,
|
|
"eval_steps_per_second": 17.274,
|
|
"step": 8
|
|
},
|
|
{
|
|
"epoch": 2.5,
|
|
"learning_rate": 2.0833333333333336e-05,
|
|
"loss": 2.7041,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 3.0,
|
|
"eval_loss": 2.68715500831604,
|
|
"eval_runtime": 0.2333,
|
|
"eval_samples_per_second": 531.614,
|
|
"eval_steps_per_second": 17.149,
|
|
"step": 12
|
|
},
|
|
{
|
|
"epoch": 4.0,
|
|
"eval_loss": 2.5978808403015137,
|
|
"eval_runtime": 0.2335,
|
|
"eval_samples_per_second": 531.073,
|
|
"eval_steps_per_second": 17.131,
|
|
"step": 16
|
|
},
|
|
{
|
|
"epoch": 5.0,
|
|
"learning_rate": 4.166666666666667e-05,
|
|
"loss": 2.5723,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 5.0,
|
|
"eval_loss": 2.4880385398864746,
|
|
"eval_runtime": 0.2334,
|
|
"eval_samples_per_second": 531.275,
|
|
"eval_steps_per_second": 17.138,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 6.0,
|
|
"eval_loss": 2.36921763420105,
|
|
"eval_runtime": 0.2335,
|
|
"eval_samples_per_second": 530.983,
|
|
"eval_steps_per_second": 17.128,
|
|
"step": 24
|
|
},
|
|
{
|
|
"epoch": 7.0,
|
|
"eval_loss": 2.251831531524658,
|
|
"eval_runtime": 0.2344,
|
|
"eval_samples_per_second": 528.969,
|
|
"eval_steps_per_second": 17.064,
|
|
"step": 28
|
|
},
|
|
{
|
|
"epoch": 7.5,
|
|
"learning_rate": 4.99926249076577e-05,
|
|
"loss": 2.2841,
|
|
"step": 30
|
|
},
|
|
{
|
|
"epoch": 8.0,
|
|
"eval_loss": 2.150097370147705,
|
|
"eval_runtime": 0.2345,
|
|
"eval_samples_per_second": 528.895,
|
|
"eval_steps_per_second": 17.061,
|
|
"step": 32
|
|
},
|
|
{
|
|
"epoch": 9.0,
|
|
"eval_loss": 2.066693067550659,
|
|
"eval_runtime": 0.2347,
|
|
"eval_samples_per_second": 528.262,
|
|
"eval_steps_per_second": 17.041,
|
|
"step": 36
|
|
},
|
|
{
|
|
"epoch": 10.0,
|
|
"learning_rate": 4.9947570655942796e-05,
|
|
"loss": 2.0314,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 10.0,
|
|
"eval_loss": 1.997233271598816,
|
|
"eval_runtime": 0.2331,
|
|
"eval_samples_per_second": 531.901,
|
|
"eval_steps_per_second": 17.158,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 11.0,
|
|
"eval_loss": 1.942152500152588,
|
|
"eval_runtime": 0.2343,
|
|
"eval_samples_per_second": 529.33,
|
|
"eval_steps_per_second": 17.075,
|
|
"step": 44
|
|
},
|
|
{
|
|
"epoch": 12.0,
|
|
"eval_loss": 1.8860777616500854,
|
|
"eval_runtime": 0.2352,
|
|
"eval_samples_per_second": 527.13,
|
|
"eval_steps_per_second": 17.004,
|
|
"step": 48
|
|
},
|
|
{
|
|
"epoch": 12.5,
|
|
"learning_rate": 4.986163316987876e-05,
|
|
"loss": 1.8417,
|
|
"step": 50
|
|
},
|
|
{
|
|
"epoch": 13.0,
|
|
"eval_loss": 1.8354158401489258,
|
|
"eval_runtime": 0.2335,
|
|
"eval_samples_per_second": 530.949,
|
|
"eval_steps_per_second": 17.127,
|
|
"step": 52
|
|
},
|
|
{
|
|
"epoch": 14.0,
|
|
"eval_loss": 1.7908005714416504,
|
|
"eval_runtime": 0.2353,
|
|
"eval_samples_per_second": 526.914,
|
|
"eval_steps_per_second": 16.997,
|
|
"step": 56
|
|
},
|
|
{
|
|
"epoch": 15.0,
|
|
"learning_rate": 4.9734953280908904e-05,
|
|
"loss": 1.7039,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 15.0,
|
|
"eval_loss": 1.749639868736267,
|
|
"eval_runtime": 0.2355,
|
|
"eval_samples_per_second": 526.443,
|
|
"eval_steps_per_second": 16.982,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 16.0,
|
|
"eval_loss": 1.7116143703460693,
|
|
"eval_runtime": 0.2351,
|
|
"eval_samples_per_second": 527.341,
|
|
"eval_steps_per_second": 17.011,
|
|
"step": 64
|
|
},
|
|
{
|
|
"epoch": 17.0,
|
|
"eval_loss": 1.6750673055648804,
|
|
"eval_runtime": 0.2344,
|
|
"eval_samples_per_second": 529.063,
|
|
"eval_steps_per_second": 17.067,
|
|
"step": 68
|
|
},
|
|
{
|
|
"epoch": 17.5,
|
|
"learning_rate": 4.956773858774731e-05,
|
|
"loss": 1.6139,
|
|
"step": 70
|
|
},
|
|
{
|
|
"epoch": 18.0,
|
|
"eval_loss": 1.6405935287475586,
|
|
"eval_runtime": 0.2334,
|
|
"eval_samples_per_second": 531.186,
|
|
"eval_steps_per_second": 17.135,
|
|
"step": 72
|
|
},
|
|
{
|
|
"epoch": 19.0,
|
|
"eval_loss": 1.6075141429901123,
|
|
"eval_runtime": 0.2335,
|
|
"eval_samples_per_second": 531.0,
|
|
"eval_steps_per_second": 17.129,
|
|
"step": 76
|
|
},
|
|
{
|
|
"epoch": 20.0,
|
|
"learning_rate": 4.936026311617316e-05,
|
|
"loss": 1.4994,
|
|
"step": 80
|
|
},
|
|
{
|
|
"epoch": 20.0,
|
|
"eval_loss": 1.5726251602172852,
|
|
"eval_runtime": 0.2335,
|
|
"eval_samples_per_second": 531.099,
|
|
"eval_steps_per_second": 17.132,
|
|
"step": 80
|
|
},
|
|
{
|
|
"epoch": 21.0,
|
|
"eval_loss": 1.5358107089996338,
|
|
"eval_runtime": 0.2341,
|
|
"eval_samples_per_second": 529.685,
|
|
"eval_steps_per_second": 17.087,
|
|
"step": 84
|
|
},
|
|
{
|
|
"epoch": 22.0,
|
|
"eval_loss": 1.5035761594772339,
|
|
"eval_runtime": 0.2352,
|
|
"eval_samples_per_second": 527.207,
|
|
"eval_steps_per_second": 17.007,
|
|
"step": 88
|
|
},
|
|
{
|
|
"epoch": 22.5,
|
|
"learning_rate": 4.9112866869966475e-05,
|
|
"loss": 1.3775,
|
|
"step": 90
|
|
},
|
|
{
|
|
"epoch": 23.0,
|
|
"eval_loss": 1.4747382402420044,
|
|
"eval_runtime": 0.234,
|
|
"eval_samples_per_second": 529.824,
|
|
"eval_steps_per_second": 17.091,
|
|
"step": 92
|
|
},
|
|
{
|
|
"epoch": 24.0,
|
|
"eval_loss": 1.4462404251098633,
|
|
"eval_runtime": 0.235,
|
|
"eval_samples_per_second": 527.752,
|
|
"eval_steps_per_second": 17.024,
|
|
"step": 96
|
|
},
|
|
{
|
|
"epoch": 25.0,
|
|
"learning_rate": 4.882595527372152e-05,
|
|
"loss": 1.3322,
|
|
"step": 100
|
|
},
|
|
{
|
|
"epoch": 25.0,
|
|
"eval_loss": 1.4174844026565552,
|
|
"eval_runtime": 0.2338,
|
|
"eval_samples_per_second": 530.293,
|
|
"eval_steps_per_second": 17.106,
|
|
"step": 100
|
|
},
|
|
{
|
|
"epoch": 26.0,
|
|
"eval_loss": 1.3916016817092896,
|
|
"eval_runtime": 0.2345,
|
|
"eval_samples_per_second": 528.89,
|
|
"eval_steps_per_second": 17.061,
|
|
"step": 104
|
|
},
|
|
{
|
|
"epoch": 27.0,
|
|
"eval_loss": 1.3674062490463257,
|
|
"eval_runtime": 0.2346,
|
|
"eval_samples_per_second": 528.576,
|
|
"eval_steps_per_second": 17.051,
|
|
"step": 108
|
|
},
|
|
{
|
|
"epoch": 27.5,
|
|
"learning_rate": 4.8499998508450664e-05,
|
|
"loss": 1.2736,
|
|
"step": 110
|
|
},
|
|
{
|
|
"epoch": 28.0,
|
|
"eval_loss": 1.3363479375839233,
|
|
"eval_runtime": 0.2343,
|
|
"eval_samples_per_second": 529.219,
|
|
"eval_steps_per_second": 17.072,
|
|
"step": 112
|
|
},
|
|
{
|
|
"epoch": 29.0,
|
|
"eval_loss": 1.3069173097610474,
|
|
"eval_runtime": 0.236,
|
|
"eval_samples_per_second": 525.476,
|
|
"eval_steps_per_second": 16.951,
|
|
"step": 116
|
|
},
|
|
{
|
|
"epoch": 30.0,
|
|
"learning_rate": 4.813553074106761e-05,
|
|
"loss": 1.159,
|
|
"step": 120
|
|
},
|
|
{
|
|
"epoch": 30.0,
|
|
"eval_loss": 1.2828550338745117,
|
|
"eval_runtime": 0.2345,
|
|
"eval_samples_per_second": 528.692,
|
|
"eval_steps_per_second": 17.055,
|
|
"step": 120
|
|
},
|
|
{
|
|
"epoch": 31.0,
|
|
"eval_loss": 1.258720874786377,
|
|
"eval_runtime": 0.2351,
|
|
"eval_samples_per_second": 527.334,
|
|
"eval_steps_per_second": 17.011,
|
|
"step": 124
|
|
},
|
|
{
|
|
"epoch": 32.0,
|
|
"eval_loss": 1.2361079454421997,
|
|
"eval_runtime": 0.2335,
|
|
"eval_samples_per_second": 531.011,
|
|
"eval_steps_per_second": 17.129,
|
|
"step": 128
|
|
},
|
|
{
|
|
"epoch": 32.5,
|
|
"learning_rate": 4.773314924901281e-05,
|
|
"loss": 1.1213,
|
|
"step": 130
|
|
},
|
|
{
|
|
"epoch": 33.0,
|
|
"eval_loss": 1.2099283933639526,
|
|
"eval_runtime": 0.2372,
|
|
"eval_samples_per_second": 522.816,
|
|
"eval_steps_per_second": 16.865,
|
|
"step": 132
|
|
},
|
|
{
|
|
"epoch": 34.0,
|
|
"eval_loss": 1.181950569152832,
|
|
"eval_runtime": 0.2365,
|
|
"eval_samples_per_second": 524.361,
|
|
"eval_steps_per_second": 16.915,
|
|
"step": 136
|
|
},
|
|
{
|
|
"epoch": 35.0,
|
|
"learning_rate": 4.7293513441455364e-05,
|
|
"loss": 1.0474,
|
|
"step": 140
|
|
},
|
|
{
|
|
"epoch": 35.0,
|
|
"eval_loss": 1.1552761793136597,
|
|
"eval_runtime": 0.2346,
|
|
"eval_samples_per_second": 528.61,
|
|
"eval_steps_per_second": 17.052,
|
|
"step": 140
|
|
},
|
|
{
|
|
"epoch": 36.0,
|
|
"eval_loss": 1.1339298486709595,
|
|
"eval_runtime": 0.2348,
|
|
"eval_samples_per_second": 528.056,
|
|
"eval_steps_per_second": 17.034,
|
|
"step": 144
|
|
},
|
|
{
|
|
"epoch": 37.0,
|
|
"eval_loss": 1.1119135618209839,
|
|
"eval_runtime": 0.2347,
|
|
"eval_samples_per_second": 528.309,
|
|
"eval_steps_per_second": 17.042,
|
|
"step": 148
|
|
},
|
|
{
|
|
"epoch": 37.5,
|
|
"learning_rate": 4.6817343778675614e-05,
|
|
"loss": 0.9878,
|
|
"step": 150
|
|
},
|
|
{
|
|
"epoch": 38.0,
|
|
"eval_loss": 1.0923570394515991,
|
|
"eval_runtime": 0.2346,
|
|
"eval_samples_per_second": 528.65,
|
|
"eval_steps_per_second": 17.053,
|
|
"step": 152
|
|
},
|
|
{
|
|
"epoch": 39.0,
|
|
"eval_loss": 1.0679152011871338,
|
|
"eval_runtime": 0.2352,
|
|
"eval_samples_per_second": 527.178,
|
|
"eval_steps_per_second": 17.006,
|
|
"step": 156
|
|
},
|
|
{
|
|
"epoch": 40.0,
|
|
"learning_rate": 4.630542059139924e-05,
|
|
"loss": 0.9045,
|
|
"step": 160
|
|
},
|
|
{
|
|
"epoch": 40.0,
|
|
"eval_loss": 1.0474611520767212,
|
|
"eval_runtime": 0.2341,
|
|
"eval_samples_per_second": 529.596,
|
|
"eval_steps_per_second": 17.084,
|
|
"step": 160
|
|
},
|
|
{
|
|
"epoch": 41.0,
|
|
"eval_loss": 1.025860071182251,
|
|
"eval_runtime": 0.2369,
|
|
"eval_samples_per_second": 523.338,
|
|
"eval_steps_per_second": 16.882,
|
|
"step": 164
|
|
},
|
|
{
|
|
"epoch": 42.0,
|
|
"eval_loss": 1.0022412538528442,
|
|
"eval_runtime": 0.2349,
|
|
"eval_samples_per_second": 527.956,
|
|
"eval_steps_per_second": 17.031,
|
|
"step": 168
|
|
},
|
|
{
|
|
"epoch": 42.5,
|
|
"learning_rate": 4.57585828020176e-05,
|
|
"loss": 0.8945,
|
|
"step": 170
|
|
},
|
|
{
|
|
"epoch": 43.0,
|
|
"eval_loss": 0.9760667085647583,
|
|
"eval_runtime": 0.2357,
|
|
"eval_samples_per_second": 526.13,
|
|
"eval_steps_per_second": 16.972,
|
|
"step": 172
|
|
},
|
|
{
|
|
"epoch": 44.0,
|
|
"eval_loss": 0.9489535689353943,
|
|
"eval_runtime": 0.2361,
|
|
"eval_samples_per_second": 525.114,
|
|
"eval_steps_per_second": 16.939,
|
|
"step": 176
|
|
},
|
|
{
|
|
"epoch": 45.0,
|
|
"learning_rate": 4.517772654979023e-05,
|
|
"loss": 0.8221,
|
|
"step": 180
|
|
},
|
|
{
|
|
"epoch": 45.0,
|
|
"eval_loss": 0.9255648255348206,
|
|
"eval_runtime": 0.2337,
|
|
"eval_samples_per_second": 530.573,
|
|
"eval_steps_per_second": 17.115,
|
|
"step": 180
|
|
},
|
|
{
|
|
"epoch": 46.0,
|
|
"eval_loss": 0.9017163515090942,
|
|
"eval_runtime": 0.2345,
|
|
"eval_samples_per_second": 528.683,
|
|
"eval_steps_per_second": 17.054,
|
|
"step": 184
|
|
},
|
|
{
|
|
"epoch": 47.0,
|
|
"eval_loss": 0.8839714527130127,
|
|
"eval_runtime": 0.2356,
|
|
"eval_samples_per_second": 526.369,
|
|
"eval_steps_per_second": 16.98,
|
|
"step": 188
|
|
},
|
|
{
|
|
"epoch": 47.5,
|
|
"learning_rate": 4.4563803722282074e-05,
|
|
"loss": 0.7435,
|
|
"step": 190
|
|
},
|
|
{
|
|
"epoch": 48.0,
|
|
"eval_loss": 0.862977921962738,
|
|
"eval_runtime": 0.2349,
|
|
"eval_samples_per_second": 527.84,
|
|
"eval_steps_per_second": 17.027,
|
|
"step": 192
|
|
},
|
|
{
|
|
"epoch": 49.0,
|
|
"eval_loss": 0.8371307849884033,
|
|
"eval_runtime": 0.2548,
|
|
"eval_samples_per_second": 486.701,
|
|
"eval_steps_per_second": 15.7,
|
|
"step": 196
|
|
},
|
|
{
|
|
"epoch": 50.0,
|
|
"learning_rate": 4.391782039544238e-05,
|
|
"loss": 0.6599,
|
|
"step": 200
|
|
},
|
|
{
|
|
"epoch": 50.0,
|
|
"eval_loss": 0.8110985159873962,
|
|
"eval_runtime": 0.2441,
|
|
"eval_samples_per_second": 507.908,
|
|
"eval_steps_per_second": 16.384,
|
|
"step": 200
|
|
},
|
|
{
|
|
"epoch": 51.0,
|
|
"eval_loss": 0.7943972945213318,
|
|
"eval_runtime": 0.2357,
|
|
"eval_samples_per_second": 526.204,
|
|
"eval_steps_per_second": 16.974,
|
|
"step": 204
|
|
},
|
|
{
|
|
"epoch": 52.0,
|
|
"eval_loss": 0.7772865295410156,
|
|
"eval_runtime": 0.2334,
|
|
"eval_samples_per_second": 531.309,
|
|
"eval_steps_per_second": 17.139,
|
|
"step": 208
|
|
},
|
|
{
|
|
"epoch": 52.5,
|
|
"learning_rate": 4.324083518488151e-05,
|
|
"loss": 0.6373,
|
|
"step": 210
|
|
},
|
|
{
|
|
"epoch": 53.0,
|
|
"eval_loss": 0.752699077129364,
|
|
"eval_runtime": 0.2416,
|
|
"eval_samples_per_second": 513.247,
|
|
"eval_steps_per_second": 16.556,
|
|
"step": 212
|
|
},
|
|
{
|
|
"epoch": 54.0,
|
|
"eval_loss": 0.7254632711410522,
|
|
"eval_runtime": 0.2507,
|
|
"eval_samples_per_second": 494.527,
|
|
"eval_steps_per_second": 15.952,
|
|
"step": 216
|
|
},
|
|
{
|
|
"epoch": 55.0,
|
|
"learning_rate": 4.253395751104748e-05,
|
|
"loss": 0.5852,
|
|
"step": 220
|
|
},
|
|
{
|
|
"epoch": 55.0,
|
|
"eval_loss": 0.7008358836174011,
|
|
"eval_runtime": 0.2358,
|
|
"eval_samples_per_second": 525.802,
|
|
"eval_steps_per_second": 16.961,
|
|
"step": 220
|
|
},
|
|
{
|
|
"epoch": 56.0,
|
|
"eval_loss": 0.6849360466003418,
|
|
"eval_runtime": 0.2335,
|
|
"eval_samples_per_second": 531.087,
|
|
"eval_steps_per_second": 17.132,
|
|
"step": 224
|
|
},
|
|
{
|
|
"epoch": 57.0,
|
|
"eval_loss": 0.6599498987197876,
|
|
"eval_runtime": 0.2351,
|
|
"eval_samples_per_second": 527.445,
|
|
"eval_steps_per_second": 17.014,
|
|
"step": 228
|
|
},
|
|
{
|
|
"epoch": 57.5,
|
|
"learning_rate": 4.179834578114531e-05,
|
|
"loss": 0.5407,
|
|
"step": 230
|
|
},
|
|
{
|
|
"epoch": 58.0,
|
|
"eval_loss": 0.6415725350379944,
|
|
"eval_runtime": 0.2347,
|
|
"eval_samples_per_second": 528.248,
|
|
"eval_steps_per_second": 17.04,
|
|
"step": 232
|
|
},
|
|
{
|
|
"epoch": 59.0,
|
|
"eval_loss": 0.6161048412322998,
|
|
"eval_runtime": 0.238,
|
|
"eval_samples_per_second": 521.069,
|
|
"eval_steps_per_second": 16.809,
|
|
"step": 236
|
|
},
|
|
{
|
|
"epoch": 60.0,
|
|
"learning_rate": 4.10352054907785e-05,
|
|
"loss": 0.49,
|
|
"step": 240
|
|
},
|
|
{
|
|
"epoch": 60.0,
|
|
"eval_loss": 0.5967058539390564,
|
|
"eval_runtime": 0.2359,
|
|
"eval_samples_per_second": 525.75,
|
|
"eval_steps_per_second": 16.96,
|
|
"step": 240
|
|
},
|
|
{
|
|
"epoch": 61.0,
|
|
"eval_loss": 0.575340747833252,
|
|
"eval_runtime": 0.2518,
|
|
"eval_samples_per_second": 492.373,
|
|
"eval_steps_per_second": 15.883,
|
|
"step": 244
|
|
},
|
|
{
|
|
"epoch": 62.0,
|
|
"eval_loss": 0.5547025799751282,
|
|
"eval_runtime": 0.2427,
|
|
"eval_samples_per_second": 510.835,
|
|
"eval_steps_per_second": 16.479,
|
|
"step": 248
|
|
},
|
|
{
|
|
"epoch": 62.5,
|
|
"learning_rate": 4.0245787248423614e-05,
|
|
"loss": 0.4613,
|
|
"step": 250
|
|
},
|
|
{
|
|
"epoch": 63.0,
|
|
"eval_loss": 0.5388721227645874,
|
|
"eval_runtime": 0.2452,
|
|
"eval_samples_per_second": 505.648,
|
|
"eval_steps_per_second": 16.311,
|
|
"step": 252
|
|
},
|
|
{
|
|
"epoch": 64.0,
|
|
"eval_loss": 0.5245070457458496,
|
|
"eval_runtime": 0.237,
|
|
"eval_samples_per_second": 523.269,
|
|
"eval_steps_per_second": 16.88,
|
|
"step": 256
|
|
},
|
|
{
|
|
"epoch": 65.0,
|
|
"learning_rate": 3.951391154534415e-05,
|
|
"loss": 0.3908,
|
|
"step": 260
|
|
},
|
|
{
|
|
"epoch": 65.0,
|
|
"eval_loss": 0.5081003308296204,
|
|
"eval_runtime": 0.2408,
|
|
"eval_samples_per_second": 514.854,
|
|
"eval_steps_per_second": 16.608,
|
|
"step": 260
|
|
},
|
|
{
|
|
"epoch": 66.0,
|
|
"eval_loss": 0.49597272276878357,
|
|
"eval_runtime": 0.2384,
|
|
"eval_samples_per_second": 520.069,
|
|
"eval_steps_per_second": 16.776,
|
|
"step": 264
|
|
},
|
|
{
|
|
"epoch": 67.0,
|
|
"eval_loss": 0.4858863651752472,
|
|
"eval_runtime": 0.2534,
|
|
"eval_samples_per_second": 489.437,
|
|
"eval_steps_per_second": 15.788,
|
|
"step": 268
|
|
},
|
|
{
|
|
"epoch": 67.5,
|
|
"learning_rate": 3.8678163144893006e-05,
|
|
"loss": 0.3568,
|
|
"step": 270
|
|
},
|
|
{
|
|
"epoch": 68.0,
|
|
"eval_loss": 0.4625568389892578,
|
|
"eval_runtime": 0.2363,
|
|
"eval_samples_per_second": 524.849,
|
|
"eval_steps_per_second": 16.931,
|
|
"step": 272
|
|
},
|
|
{
|
|
"epoch": 69.0,
|
|
"eval_loss": 0.44423165917396545,
|
|
"eval_runtime": 0.2347,
|
|
"eval_samples_per_second": 528.262,
|
|
"eval_steps_per_second": 17.041,
|
|
"step": 276
|
|
},
|
|
{
|
|
"epoch": 70.0,
|
|
"learning_rate": 3.7819999433667504e-05,
|
|
"loss": 0.3277,
|
|
"step": 280
|
|
},
|
|
{
|
|
"epoch": 70.0,
|
|
"eval_loss": 0.4274282157421112,
|
|
"eval_runtime": 0.2349,
|
|
"eval_samples_per_second": 527.857,
|
|
"eval_steps_per_second": 17.028,
|
|
"step": 280
|
|
},
|
|
{
|
|
"epoch": 71.0,
|
|
"eval_loss": 0.41012296080589294,
|
|
"eval_runtime": 0.2353,
|
|
"eval_samples_per_second": 527.096,
|
|
"eval_steps_per_second": 17.003,
|
|
"step": 284
|
|
},
|
|
{
|
|
"epoch": 72.0,
|
|
"eval_loss": 0.39118534326553345,
|
|
"eval_runtime": 0.2369,
|
|
"eval_samples_per_second": 523.337,
|
|
"eval_steps_per_second": 16.882,
|
|
"step": 288
|
|
},
|
|
{
|
|
"epoch": 72.5,
|
|
"learning_rate": 3.694082674130991e-05,
|
|
"loss": 0.2979,
|
|
"step": 290
|
|
},
|
|
{
|
|
"epoch": 73.0,
|
|
"eval_loss": 0.3788595497608185,
|
|
"eval_runtime": 0.2425,
|
|
"eval_samples_per_second": 511.42,
|
|
"eval_steps_per_second": 16.497,
|
|
"step": 292
|
|
},
|
|
{
|
|
"epoch": 74.0,
|
|
"eval_loss": 0.35853779315948486,
|
|
"eval_runtime": 0.2441,
|
|
"eval_samples_per_second": 508.031,
|
|
"eval_steps_per_second": 16.388,
|
|
"step": 296
|
|
},
|
|
{
|
|
"epoch": 75.0,
|
|
"learning_rate": 3.604208582626964e-05,
|
|
"loss": 0.2716,
|
|
"step": 300
|
|
},
|
|
{
|
|
"epoch": 75.0,
|
|
"eval_loss": 0.34690794348716736,
|
|
"eval_runtime": 0.2473,
|
|
"eval_samples_per_second": 501.338,
|
|
"eval_steps_per_second": 16.172,
|
|
"step": 300
|
|
},
|
|
{
|
|
"epoch": 76.0,
|
|
"eval_loss": 0.33692988753318787,
|
|
"eval_runtime": 0.2511,
|
|
"eval_samples_per_second": 493.843,
|
|
"eval_steps_per_second": 15.93,
|
|
"step": 304
|
|
},
|
|
{
|
|
"epoch": 77.0,
|
|
"eval_loss": 0.3223101794719696,
|
|
"eval_runtime": 0.2687,
|
|
"eval_samples_per_second": 461.53,
|
|
"eval_steps_per_second": 14.888,
|
|
"step": 308
|
|
},
|
|
{
|
|
"epoch": 77.5,
|
|
"learning_rate": 3.512524951473709e-05,
|
|
"loss": 0.2413,
|
|
"step": 310
|
|
},
|
|
{
|
|
"epoch": 78.0,
|
|
"eval_loss": 0.3038597106933594,
|
|
"eval_runtime": 0.275,
|
|
"eval_samples_per_second": 450.923,
|
|
"eval_steps_per_second": 14.546,
|
|
"step": 312
|
|
},
|
|
{
|
|
"epoch": 79.0,
|
|
"eval_loss": 0.2942875325679779,
|
|
"eval_runtime": 0.2331,
|
|
"eval_samples_per_second": 532.01,
|
|
"eval_steps_per_second": 17.162,
|
|
"step": 316
|
|
},
|
|
{
|
|
"epoch": 80.0,
|
|
"learning_rate": 3.4285866497155414e-05,
|
|
"loss": 0.2053,
|
|
"step": 320
|
|
},
|
|
{
|
|
"epoch": 80.0,
|
|
"eval_loss": 0.28482967615127563,
|
|
"eval_runtime": 0.2355,
|
|
"eval_samples_per_second": 526.629,
|
|
"eval_steps_per_second": 16.988,
|
|
"step": 320
|
|
},
|
|
{
|
|
"epoch": 81.0,
|
|
"eval_loss": 0.26999109983444214,
|
|
"eval_runtime": 0.2343,
|
|
"eval_samples_per_second": 529.13,
|
|
"eval_steps_per_second": 17.069,
|
|
"step": 324
|
|
},
|
|
{
|
|
"epoch": 82.0,
|
|
"eval_loss": 0.25828245282173157,
|
|
"eval_runtime": 0.2349,
|
|
"eval_samples_per_second": 527.929,
|
|
"eval_steps_per_second": 17.03,
|
|
"step": 328
|
|
},
|
|
{
|
|
"epoch": 82.5,
|
|
"learning_rate": 3.333881079127052e-05,
|
|
"loss": 0.1794,
|
|
"step": 330
|
|
},
|
|
{
|
|
"epoch": 83.0,
|
|
"eval_loss": 0.24674412608146667,
|
|
"eval_runtime": 0.2358,
|
|
"eval_samples_per_second": 525.869,
|
|
"eval_steps_per_second": 16.964,
|
|
"step": 332
|
|
},
|
|
{
|
|
"epoch": 84.0,
|
|
"eval_loss": 0.23660407960414886,
|
|
"eval_runtime": 0.2353,
|
|
"eval_samples_per_second": 527.022,
|
|
"eval_steps_per_second": 17.001,
|
|
"step": 336
|
|
},
|
|
{
|
|
"epoch": 85.0,
|
|
"learning_rate": 3.2378089724518465e-05,
|
|
"loss": 0.1708,
|
|
"step": 340
|
|
},
|
|
{
|
|
"epoch": 85.0,
|
|
"eval_loss": 0.2259848266839981,
|
|
"eval_runtime": 0.234,
|
|
"eval_samples_per_second": 530.003,
|
|
"eval_steps_per_second": 17.097,
|
|
"step": 340
|
|
},
|
|
{
|
|
"epoch": 86.0,
|
|
"eval_loss": 0.2151513248682022,
|
|
"eval_runtime": 0.2358,
|
|
"eval_samples_per_second": 525.8,
|
|
"eval_steps_per_second": 16.961,
|
|
"step": 344
|
|
},
|
|
{
|
|
"epoch": 87.0,
|
|
"eval_loss": 0.203897625207901,
|
|
"eval_runtime": 0.2341,
|
|
"eval_samples_per_second": 529.731,
|
|
"eval_steps_per_second": 17.088,
|
|
"step": 348
|
|
},
|
|
{
|
|
"epoch": 87.5,
|
|
"learning_rate": 3.1405277694064305e-05,
|
|
"loss": 0.1371,
|
|
"step": 350
|
|
},
|
|
{
|
|
"epoch": 88.0,
|
|
"eval_loss": 0.20068475604057312,
|
|
"eval_runtime": 0.2339,
|
|
"eval_samples_per_second": 530.128,
|
|
"eval_steps_per_second": 17.101,
|
|
"step": 352
|
|
},
|
|
{
|
|
"epoch": 89.0,
|
|
"eval_loss": 0.1907888799905777,
|
|
"eval_runtime": 0.2319,
|
|
"eval_samples_per_second": 534.614,
|
|
"eval_steps_per_second": 17.246,
|
|
"step": 356
|
|
},
|
|
{
|
|
"epoch": 90.0,
|
|
"learning_rate": 3.052072634104116e-05,
|
|
"loss": 0.1305,
|
|
"step": 360
|
|
},
|
|
{
|
|
"epoch": 90.0,
|
|
"eval_loss": 0.18260663747787476,
|
|
"eval_runtime": 0.2343,
|
|
"eval_samples_per_second": 529.305,
|
|
"eval_steps_per_second": 17.074,
|
|
"step": 360
|
|
},
|
|
{
|
|
"epoch": 91.0,
|
|
"eval_loss": 0.17595653235912323,
|
|
"eval_runtime": 0.2342,
|
|
"eval_samples_per_second": 529.488,
|
|
"eval_steps_per_second": 17.08,
|
|
"step": 364
|
|
},
|
|
{
|
|
"epoch": 92.0,
|
|
"eval_loss": 0.16729047894477844,
|
|
"eval_runtime": 0.2363,
|
|
"eval_samples_per_second": 524.736,
|
|
"eval_steps_per_second": 16.927,
|
|
"step": 368
|
|
},
|
|
{
|
|
"epoch": 92.5,
|
|
"learning_rate": 2.952934780370694e-05,
|
|
"loss": 0.1133,
|
|
"step": 370
|
|
},
|
|
{
|
|
"epoch": 93.0,
|
|
"eval_loss": 0.15900956094264984,
|
|
"eval_runtime": 0.2361,
|
|
"eval_samples_per_second": 525.156,
|
|
"eval_steps_per_second": 16.941,
|
|
"step": 372
|
|
},
|
|
{
|
|
"epoch": 94.0,
|
|
"eval_loss": 0.14826063811779022,
|
|
"eval_runtime": 0.2338,
|
|
"eval_samples_per_second": 530.441,
|
|
"eval_steps_per_second": 17.111,
|
|
"step": 376
|
|
},
|
|
{
|
|
"epoch": 95.0,
|
|
"learning_rate": 2.853054672450437e-05,
|
|
"loss": 0.1111,
|
|
"step": 380
|
|
},
|
|
{
|
|
"epoch": 95.0,
|
|
"eval_loss": 0.139539897441864,
|
|
"eval_runtime": 0.2345,
|
|
"eval_samples_per_second": 528.849,
|
|
"eval_steps_per_second": 17.06,
|
|
"step": 380
|
|
},
|
|
{
|
|
"epoch": 96.0,
|
|
"eval_loss": 0.13384650647640228,
|
|
"eval_runtime": 0.2368,
|
|
"eval_samples_per_second": 523.583,
|
|
"eval_steps_per_second": 16.89,
|
|
"step": 384
|
|
},
|
|
{
|
|
"epoch": 97.0,
|
|
"eval_loss": 0.1287873536348343,
|
|
"eval_runtime": 0.2349,
|
|
"eval_samples_per_second": 527.829,
|
|
"eval_steps_per_second": 17.027,
|
|
"step": 388
|
|
},
|
|
{
|
|
"epoch": 97.5,
|
|
"learning_rate": 2.7525959904833952e-05,
|
|
"loss": 0.0971,
|
|
"step": 390
|
|
},
|
|
{
|
|
"epoch": 98.0,
|
|
"eval_loss": 0.12157784402370453,
|
|
"eval_runtime": 0.2318,
|
|
"eval_samples_per_second": 535.023,
|
|
"eval_steps_per_second": 17.259,
|
|
"step": 392
|
|
},
|
|
{
|
|
"epoch": 99.0,
|
|
"eval_loss": 0.11712194234132767,
|
|
"eval_runtime": 0.2353,
|
|
"eval_samples_per_second": 527.025,
|
|
"eval_steps_per_second": 17.001,
|
|
"step": 396
|
|
},
|
|
{
|
|
"epoch": 100.0,
|
|
"learning_rate": 2.6517233627571858e-05,
|
|
"loss": 0.0809,
|
|
"step": 400
|
|
},
|
|
{
|
|
"epoch": 100.0,
|
|
"eval_loss": 0.10868225991725922,
|
|
"eval_runtime": 0.235,
|
|
"eval_samples_per_second": 527.572,
|
|
"eval_steps_per_second": 17.018,
|
|
"step": 400
|
|
},
|
|
{
|
|
"epoch": 101.0,
|
|
"eval_loss": 0.10526269674301147,
|
|
"eval_runtime": 0.2349,
|
|
"eval_samples_per_second": 527.774,
|
|
"eval_steps_per_second": 17.025,
|
|
"step": 404
|
|
},
|
|
{
|
|
"epoch": 102.0,
|
|
"eval_loss": 0.10188782960176468,
|
|
"eval_runtime": 0.236,
|
|
"eval_samples_per_second": 525.459,
|
|
"eval_steps_per_second": 16.95,
|
|
"step": 408
|
|
},
|
|
{
|
|
"epoch": 102.5,
|
|
"learning_rate": 2.550602095919722e-05,
|
|
"loss": 0.0737,
|
|
"step": 410
|
|
},
|
|
{
|
|
"epoch": 103.0,
|
|
"eval_loss": 0.09672608226537704,
|
|
"eval_runtime": 0.2341,
|
|
"eval_samples_per_second": 529.772,
|
|
"eval_steps_per_second": 17.089,
|
|
"step": 412
|
|
},
|
|
{
|
|
"epoch": 104.0,
|
|
"eval_loss": 0.09119828790426254,
|
|
"eval_runtime": 0.234,
|
|
"eval_samples_per_second": 529.835,
|
|
"eval_steps_per_second": 17.091,
|
|
"step": 416
|
|
},
|
|
{
|
|
"epoch": 105.0,
|
|
"learning_rate": 2.4493979040802785e-05,
|
|
"loss": 0.0673,
|
|
"step": 420
|
|
},
|
|
{
|
|
"epoch": 105.0,
|
|
"eval_loss": 0.08706870675086975,
|
|
"eval_runtime": 0.2339,
|
|
"eval_samples_per_second": 530.032,
|
|
"eval_steps_per_second": 17.098,
|
|
"step": 420
|
|
},
|
|
{
|
|
"epoch": 106.0,
|
|
"eval_loss": 0.08334846049547195,
|
|
"eval_runtime": 0.2337,
|
|
"eval_samples_per_second": 530.574,
|
|
"eval_steps_per_second": 17.115,
|
|
"step": 424
|
|
},
|
|
{
|
|
"epoch": 107.0,
|
|
"eval_loss": 0.07730688899755478,
|
|
"eval_runtime": 0.2367,
|
|
"eval_samples_per_second": 523.79,
|
|
"eval_steps_per_second": 16.896,
|
|
"step": 428
|
|
},
|
|
{
|
|
"epoch": 107.5,
|
|
"learning_rate": 2.348276637242814e-05,
|
|
"loss": 0.0566,
|
|
"step": 430
|
|
},
|
|
{
|
|
"epoch": 108.0,
|
|
"eval_loss": 0.07153230905532837,
|
|
"eval_runtime": 0.2353,
|
|
"eval_samples_per_second": 527.093,
|
|
"eval_steps_per_second": 17.003,
|
|
"step": 432
|
|
},
|
|
{
|
|
"epoch": 109.0,
|
|
"eval_loss": 0.06880535930395126,
|
|
"eval_runtime": 0.2366,
|
|
"eval_samples_per_second": 524.044,
|
|
"eval_steps_per_second": 16.905,
|
|
"step": 436
|
|
},
|
|
{
|
|
"epoch": 110.0,
|
|
"learning_rate": 2.247404009516605e-05,
|
|
"loss": 0.0508,
|
|
"step": 440
|
|
},
|
|
{
|
|
"epoch": 110.0,
|
|
"eval_loss": 0.0659087672829628,
|
|
"eval_runtime": 0.2328,
|
|
"eval_samples_per_second": 532.631,
|
|
"eval_steps_per_second": 17.182,
|
|
"step": 440
|
|
},
|
|
{
|
|
"epoch": 111.0,
|
|
"eval_loss": 0.06361372768878937,
|
|
"eval_runtime": 0.2362,
|
|
"eval_samples_per_second": 524.949,
|
|
"eval_steps_per_second": 16.934,
|
|
"step": 444
|
|
},
|
|
{
|
|
"epoch": 112.0,
|
|
"eval_loss": 0.06115693971514702,
|
|
"eval_runtime": 0.235,
|
|
"eval_samples_per_second": 527.577,
|
|
"eval_steps_per_second": 17.019,
|
|
"step": 448
|
|
},
|
|
{
|
|
"epoch": 112.5,
|
|
"learning_rate": 2.1469453275495636e-05,
|
|
"loss": 0.0445,
|
|
"step": 450
|
|
},
|
|
{
|
|
"epoch": 113.0,
|
|
"eval_loss": 0.05979010462760925,
|
|
"eval_runtime": 0.2344,
|
|
"eval_samples_per_second": 529.014,
|
|
"eval_steps_per_second": 17.065,
|
|
"step": 452
|
|
},
|
|
{
|
|
"epoch": 114.0,
|
|
"eval_loss": 0.058374565094709396,
|
|
"eval_runtime": 0.2347,
|
|
"eval_samples_per_second": 528.387,
|
|
"eval_steps_per_second": 17.045,
|
|
"step": 456
|
|
},
|
|
{
|
|
"epoch": 115.0,
|
|
"learning_rate": 2.047065219629306e-05,
|
|
"loss": 0.0432,
|
|
"step": 460
|
|
},
|
|
{
|
|
"epoch": 115.0,
|
|
"eval_loss": 0.05612092837691307,
|
|
"eval_runtime": 0.2356,
|
|
"eval_samples_per_second": 526.309,
|
|
"eval_steps_per_second": 16.978,
|
|
"step": 460
|
|
},
|
|
{
|
|
"epoch": 116.0,
|
|
"eval_loss": 0.053940750658512115,
|
|
"eval_runtime": 0.2348,
|
|
"eval_samples_per_second": 528.101,
|
|
"eval_steps_per_second": 17.036,
|
|
"step": 464
|
|
},
|
|
{
|
|
"epoch": 117.0,
|
|
"eval_loss": 0.05062006413936615,
|
|
"eval_runtime": 0.2345,
|
|
"eval_samples_per_second": 528.696,
|
|
"eval_steps_per_second": 17.055,
|
|
"step": 468
|
|
},
|
|
{
|
|
"epoch": 117.5,
|
|
"learning_rate": 1.947927365895885e-05,
|
|
"loss": 0.0389,
|
|
"step": 470
|
|
},
|
|
{
|
|
"epoch": 118.0,
|
|
"eval_loss": 0.04845296964049339,
|
|
"eval_runtime": 0.2349,
|
|
"eval_samples_per_second": 527.833,
|
|
"eval_steps_per_second": 17.027,
|
|
"step": 472
|
|
},
|
|
{
|
|
"epoch": 119.0,
|
|
"eval_loss": 0.04602435976266861,
|
|
"eval_runtime": 0.2367,
|
|
"eval_samples_per_second": 523.878,
|
|
"eval_steps_per_second": 16.899,
|
|
"step": 476
|
|
},
|
|
{
|
|
"epoch": 120.0,
|
|
"learning_rate": 1.8496942301083144e-05,
|
|
"loss": 0.0339,
|
|
"step": 480
|
|
},
|
|
{
|
|
"epoch": 120.0,
|
|
"eval_loss": 0.04395240917801857,
|
|
"eval_runtime": 0.2341,
|
|
"eval_samples_per_second": 529.751,
|
|
"eval_steps_per_second": 17.089,
|
|
"step": 480
|
|
},
|
|
{
|
|
"epoch": 121.0,
|
|
"eval_loss": 0.041246574372053146,
|
|
"eval_runtime": 0.2358,
|
|
"eval_samples_per_second": 525.885,
|
|
"eval_steps_per_second": 16.964,
|
|
"step": 484
|
|
},
|
|
{
|
|
"epoch": 122.0,
|
|
"eval_loss": 0.03897799178957939,
|
|
"eval_runtime": 0.2352,
|
|
"eval_samples_per_second": 527.121,
|
|
"eval_steps_per_second": 17.004,
|
|
"step": 488
|
|
},
|
|
{
|
|
"epoch": 122.5,
|
|
"learning_rate": 1.7525267934044643e-05,
|
|
"loss": 0.0314,
|
|
"step": 490
|
|
},
|
|
{
|
|
"epoch": 123.0,
|
|
"eval_loss": 0.037170685827732086,
|
|
"eval_runtime": 0.2342,
|
|
"eval_samples_per_second": 529.518,
|
|
"eval_steps_per_second": 17.081,
|
|
"step": 492
|
|
},
|
|
{
|
|
"epoch": 124.0,
|
|
"eval_loss": 0.035379957407712936,
|
|
"eval_runtime": 0.236,
|
|
"eval_samples_per_second": 525.526,
|
|
"eval_steps_per_second": 16.952,
|
|
"step": 496
|
|
},
|
|
{
|
|
"epoch": 125.0,
|
|
"learning_rate": 1.6565842904906153e-05,
|
|
"loss": 0.0282,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 125.0,
|
|
"eval_loss": 0.03376702591776848,
|
|
"eval_runtime": 0.2376,
|
|
"eval_samples_per_second": 521.849,
|
|
"eval_steps_per_second": 16.834,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 126.0,
|
|
"eval_loss": 0.03227021545171738,
|
|
"eval_runtime": 0.2351,
|
|
"eval_samples_per_second": 527.542,
|
|
"eval_steps_per_second": 17.017,
|
|
"step": 504
|
|
},
|
|
{
|
|
"epoch": 127.0,
|
|
"eval_loss": 0.032056521624326706,
|
|
"eval_runtime": 0.2352,
|
|
"eval_samples_per_second": 527.152,
|
|
"eval_steps_per_second": 17.005,
|
|
"step": 508
|
|
},
|
|
{
|
|
"epoch": 127.5,
|
|
"learning_rate": 1.571413350284459e-05,
|
|
"loss": 0.0263,
|
|
"step": 510
|
|
},
|
|
{
|
|
"epoch": 128.0,
|
|
"eval_loss": 0.03086697869002819,
|
|
"eval_runtime": 0.2356,
|
|
"eval_samples_per_second": 526.228,
|
|
"eval_steps_per_second": 16.975,
|
|
"step": 512
|
|
},
|
|
{
|
|
"epoch": 129.0,
|
|
"eval_loss": 0.029199883341789246,
|
|
"eval_runtime": 0.234,
|
|
"eval_samples_per_second": 529.947,
|
|
"eval_steps_per_second": 17.095,
|
|
"step": 516
|
|
},
|
|
{
|
|
"epoch": 130.0,
|
|
"learning_rate": 1.4782295160661103e-05,
|
|
"loss": 0.0227,
|
|
"step": 520
|
|
},
|
|
{
|
|
"epoch": 130.0,
|
|
"eval_loss": 0.02830178663134575,
|
|
"eval_runtime": 0.2355,
|
|
"eval_samples_per_second": 526.603,
|
|
"eval_steps_per_second": 16.987,
|
|
"step": 520
|
|
},
|
|
{
|
|
"epoch": 131.0,
|
|
"eval_loss": 0.027501724660396576,
|
|
"eval_runtime": 0.2344,
|
|
"eval_samples_per_second": 529.108,
|
|
"eval_steps_per_second": 17.068,
|
|
"step": 524
|
|
},
|
|
{
|
|
"epoch": 132.0,
|
|
"eval_loss": 0.026682091876864433,
|
|
"eval_runtime": 0.2344,
|
|
"eval_samples_per_second": 529.091,
|
|
"eval_steps_per_second": 17.067,
|
|
"step": 528
|
|
},
|
|
{
|
|
"epoch": 132.5,
|
|
"learning_rate": 1.3867201247312697e-05,
|
|
"loss": 0.0222,
|
|
"step": 530
|
|
},
|
|
{
|
|
"epoch": 133.0,
|
|
"eval_loss": 0.025750810280442238,
|
|
"eval_runtime": 0.2355,
|
|
"eval_samples_per_second": 526.44,
|
|
"eval_steps_per_second": 16.982,
|
|
"step": 532
|
|
},
|
|
{
|
|
"epoch": 134.0,
|
|
"eval_loss": 0.024244097992777824,
|
|
"eval_runtime": 0.2341,
|
|
"eval_samples_per_second": 529.638,
|
|
"eval_steps_per_second": 17.085,
|
|
"step": 536
|
|
},
|
|
{
|
|
"epoch": 135.0,
|
|
"learning_rate": 1.2970351387729873e-05,
|
|
"loss": 0.0203,
|
|
"step": 540
|
|
},
|
|
{
|
|
"epoch": 135.0,
|
|
"eval_loss": 0.023219697177410126,
|
|
"eval_runtime": 0.2354,
|
|
"eval_samples_per_second": 526.855,
|
|
"eval_steps_per_second": 16.995,
|
|
"step": 540
|
|
},
|
|
{
|
|
"epoch": 136.0,
|
|
"eval_loss": 0.022541573271155357,
|
|
"eval_runtime": 0.2361,
|
|
"eval_samples_per_second": 525.173,
|
|
"eval_steps_per_second": 16.941,
|
|
"step": 544
|
|
},
|
|
{
|
|
"epoch": 137.0,
|
|
"eval_loss": 0.021795254200696945,
|
|
"eval_runtime": 0.2337,
|
|
"eval_samples_per_second": 530.513,
|
|
"eval_steps_per_second": 17.113,
|
|
"step": 548
|
|
},
|
|
{
|
|
"epoch": 137.5,
|
|
"learning_rate": 1.2093215309105352e-05,
|
|
"loss": 0.02,
|
|
"step": 550
|
|
},
|
|
{
|
|
"epoch": 138.0,
|
|
"eval_loss": 0.021266764029860497,
|
|
"eval_runtime": 0.2348,
|
|
"eval_samples_per_second": 528.056,
|
|
"eval_steps_per_second": 17.034,
|
|
"step": 552
|
|
},
|
|
{
|
|
"epoch": 139.0,
|
|
"eval_loss": 0.02016260474920273,
|
|
"eval_runtime": 0.2431,
|
|
"eval_samples_per_second": 509.997,
|
|
"eval_steps_per_second": 16.452,
|
|
"step": 556
|
|
},
|
|
{
|
|
"epoch": 140.0,
|
|
"learning_rate": 1.1237230432354912e-05,
|
|
"loss": 0.0178,
|
|
"step": 560
|
|
},
|
|
{
|
|
"epoch": 140.0,
|
|
"eval_loss": 0.019516441971063614,
|
|
"eval_runtime": 0.2344,
|
|
"eval_samples_per_second": 529.119,
|
|
"eval_steps_per_second": 17.068,
|
|
"step": 560
|
|
},
|
|
{
|
|
"epoch": 141.0,
|
|
"eval_loss": 0.019118839874863625,
|
|
"eval_runtime": 0.2341,
|
|
"eval_samples_per_second": 529.801,
|
|
"eval_steps_per_second": 17.09,
|
|
"step": 564
|
|
},
|
|
{
|
|
"epoch": 142.0,
|
|
"eval_loss": 0.019044984132051468,
|
|
"eval_runtime": 0.2343,
|
|
"eval_samples_per_second": 529.182,
|
|
"eval_steps_per_second": 17.07,
|
|
"step": 568
|
|
},
|
|
{
|
|
"epoch": 142.5,
|
|
"learning_rate": 1.0403799516520618e-05,
|
|
"loss": 0.0173,
|
|
"step": 570
|
|
},
|
|
{
|
|
"epoch": 143.0,
|
|
"eval_loss": 0.01844731532037258,
|
|
"eval_runtime": 0.2354,
|
|
"eval_samples_per_second": 526.67,
|
|
"eval_steps_per_second": 16.989,
|
|
"step": 572
|
|
},
|
|
{
|
|
"epoch": 144.0,
|
|
"eval_loss": 0.01802230253815651,
|
|
"eval_runtime": 0.2349,
|
|
"eval_samples_per_second": 527.789,
|
|
"eval_steps_per_second": 17.025,
|
|
"step": 576
|
|
},
|
|
{
|
|
"epoch": 145.0,
|
|
"learning_rate": 9.594288359976817e-06,
|
|
"loss": 0.0162,
|
|
"step": 580
|
|
},
|
|
{
|
|
"epoch": 145.0,
|
|
"eval_loss": 0.017573414370417595,
|
|
"eval_runtime": 0.2354,
|
|
"eval_samples_per_second": 526.693,
|
|
"eval_steps_per_second": 16.99,
|
|
"step": 580
|
|
},
|
|
{
|
|
"epoch": 146.0,
|
|
"eval_loss": 0.017251133918762207,
|
|
"eval_runtime": 0.2334,
|
|
"eval_samples_per_second": 531.375,
|
|
"eval_steps_per_second": 17.141,
|
|
"step": 584
|
|
},
|
|
{
|
|
"epoch": 147.0,
|
|
"eval_loss": 0.01694837398827076,
|
|
"eval_runtime": 0.2357,
|
|
"eval_samples_per_second": 526.134,
|
|
"eval_steps_per_second": 16.972,
|
|
"step": 588
|
|
},
|
|
{
|
|
"epoch": 147.5,
|
|
"learning_rate": 8.810023562206e-06,
|
|
"loss": 0.0149,
|
|
"step": 590
|
|
},
|
|
{
|
|
"epoch": 148.0,
|
|
"eval_loss": 0.0165999922901392,
|
|
"eval_runtime": 0.2343,
|
|
"eval_samples_per_second": 529.345,
|
|
"eval_steps_per_second": 17.076,
|
|
"step": 592
|
|
},
|
|
{
|
|
"epoch": 149.0,
|
|
"eval_loss": 0.016359616070985794,
|
|
"eval_runtime": 0.2348,
|
|
"eval_samples_per_second": 528.214,
|
|
"eval_steps_per_second": 17.039,
|
|
"step": 596
|
|
},
|
|
{
|
|
"epoch": 150.0,
|
|
"learning_rate": 8.052290349812419e-06,
|
|
"loss": 0.0147,
|
|
"step": 600
|
|
},
|
|
{
|
|
"epoch": 150.0,
|
|
"eval_loss": 0.01602223329246044,
|
|
"eval_runtime": 0.2344,
|
|
"eval_samples_per_second": 528.992,
|
|
"eval_steps_per_second": 17.064,
|
|
"step": 600
|
|
},
|
|
{
|
|
"epoch": 151.0,
|
|
"eval_loss": 0.01573454588651657,
|
|
"eval_runtime": 0.234,
|
|
"eval_samples_per_second": 529.864,
|
|
"eval_steps_per_second": 17.092,
|
|
"step": 604
|
|
},
|
|
{
|
|
"epoch": 152.0,
|
|
"eval_loss": 0.015604463405907154,
|
|
"eval_runtime": 0.2348,
|
|
"eval_samples_per_second": 528.066,
|
|
"eval_steps_per_second": 17.034,
|
|
"step": 608
|
|
},
|
|
{
|
|
"epoch": 152.5,
|
|
"learning_rate": 7.3223304703363135e-06,
|
|
"loss": 0.0139,
|
|
"step": 610
|
|
},
|
|
{
|
|
"epoch": 153.0,
|
|
"eval_loss": 0.015372724272310734,
|
|
"eval_runtime": 0.2337,
|
|
"eval_samples_per_second": 530.486,
|
|
"eval_steps_per_second": 17.112,
|
|
"step": 612
|
|
},
|
|
{
|
|
"epoch": 154.0,
|
|
"eval_loss": 0.015068690292537212,
|
|
"eval_runtime": 0.2333,
|
|
"eval_samples_per_second": 531.446,
|
|
"eval_steps_per_second": 17.143,
|
|
"step": 616
|
|
},
|
|
{
|
|
"epoch": 155.0,
|
|
"learning_rate": 6.621340157319997e-06,
|
|
"loss": 0.0131,
|
|
"step": 620
|
|
},
|
|
{
|
|
"epoch": 155.0,
|
|
"eval_loss": 0.014798024669289589,
|
|
"eval_runtime": 0.2356,
|
|
"eval_samples_per_second": 526.401,
|
|
"eval_steps_per_second": 16.981,
|
|
"step": 620
|
|
},
|
|
{
|
|
"epoch": 156.0,
|
|
"eval_loss": 0.014530870132148266,
|
|
"eval_runtime": 0.2356,
|
|
"eval_samples_per_second": 526.34,
|
|
"eval_steps_per_second": 16.979,
|
|
"step": 624
|
|
},
|
|
{
|
|
"epoch": 157.0,
|
|
"eval_loss": 0.014405559748411179,
|
|
"eval_runtime": 0.2358,
|
|
"eval_samples_per_second": 525.878,
|
|
"eval_steps_per_second": 16.964,
|
|
"step": 628
|
|
},
|
|
{
|
|
"epoch": 157.5,
|
|
"learning_rate": 5.950468169960846e-06,
|
|
"loss": 0.0124,
|
|
"step": 630
|
|
},
|
|
{
|
|
"epoch": 158.0,
|
|
"eval_loss": 0.014271734282374382,
|
|
"eval_runtime": 0.2345,
|
|
"eval_samples_per_second": 528.678,
|
|
"eval_steps_per_second": 17.054,
|
|
"step": 632
|
|
},
|
|
{
|
|
"epoch": 159.0,
|
|
"eval_loss": 0.014108818024396896,
|
|
"eval_runtime": 0.2349,
|
|
"eval_samples_per_second": 527.965,
|
|
"eval_steps_per_second": 17.031,
|
|
"step": 636
|
|
},
|
|
{
|
|
"epoch": 160.0,
|
|
"learning_rate": 5.310813910563644e-06,
|
|
"loss": 0.0122,
|
|
"step": 640
|
|
},
|
|
{
|
|
"epoch": 160.0,
|
|
"eval_loss": 0.014046107418835163,
|
|
"eval_runtime": 0.2352,
|
|
"eval_samples_per_second": 527.271,
|
|
"eval_steps_per_second": 17.009,
|
|
"step": 640
|
|
},
|
|
{
|
|
"epoch": 161.0,
|
|
"eval_loss": 0.014189131557941437,
|
|
"eval_runtime": 0.235,
|
|
"eval_samples_per_second": 527.603,
|
|
"eval_steps_per_second": 17.019,
|
|
"step": 644
|
|
},
|
|
{
|
|
"epoch": 162.0,
|
|
"eval_loss": 0.014161473140120506,
|
|
"eval_runtime": 0.2362,
|
|
"eval_samples_per_second": 524.899,
|
|
"eval_steps_per_second": 16.932,
|
|
"step": 648
|
|
},
|
|
{
|
|
"epoch": 162.5,
|
|
"learning_rate": 4.703425622877239e-06,
|
|
"loss": 0.0129,
|
|
"step": 650
|
|
},
|
|
{
|
|
"epoch": 163.0,
|
|
"eval_loss": 0.014157706871628761,
|
|
"eval_runtime": 0.2339,
|
|
"eval_samples_per_second": 530.162,
|
|
"eval_steps_per_second": 17.102,
|
|
"step": 652
|
|
},
|
|
{
|
|
"epoch": 164.0,
|
|
"eval_loss": 0.013872063718736172,
|
|
"eval_runtime": 0.2365,
|
|
"eval_samples_per_second": 524.372,
|
|
"eval_steps_per_second": 16.915,
|
|
"step": 656
|
|
},
|
|
{
|
|
"epoch": 165.0,
|
|
"learning_rate": 4.129298674268225e-06,
|
|
"loss": 0.0122,
|
|
"step": 660
|
|
},
|
|
{
|
|
"epoch": 165.0,
|
|
"eval_loss": 0.013607202097773552,
|
|
"eval_runtime": 0.2332,
|
|
"eval_samples_per_second": 531.786,
|
|
"eval_steps_per_second": 17.154,
|
|
"step": 660
|
|
},
|
|
{
|
|
"epoch": 166.0,
|
|
"eval_loss": 0.013392354361712933,
|
|
"eval_runtime": 0.2346,
|
|
"eval_samples_per_second": 528.556,
|
|
"eval_steps_per_second": 17.05,
|
|
"step": 664
|
|
},
|
|
{
|
|
"epoch": 167.0,
|
|
"eval_loss": 0.013276197947561741,
|
|
"eval_runtime": 0.2356,
|
|
"eval_samples_per_second": 526.216,
|
|
"eval_steps_per_second": 16.975,
|
|
"step": 668
|
|
},
|
|
{
|
|
"epoch": 167.5,
|
|
"learning_rate": 3.5893739245465465e-06,
|
|
"loss": 0.0119,
|
|
"step": 670
|
|
},
|
|
{
|
|
"epoch": 168.0,
|
|
"eval_loss": 0.013199027627706528,
|
|
"eval_runtime": 0.2344,
|
|
"eval_samples_per_second": 529.012,
|
|
"eval_steps_per_second": 17.065,
|
|
"step": 672
|
|
},
|
|
{
|
|
"epoch": 169.0,
|
|
"eval_loss": 0.013125129044055939,
|
|
"eval_runtime": 0.2397,
|
|
"eval_samples_per_second": 517.285,
|
|
"eval_steps_per_second": 16.687,
|
|
"step": 676
|
|
},
|
|
{
|
|
"epoch": 170.0,
|
|
"learning_rate": 3.08453618411631e-06,
|
|
"loss": 0.0118,
|
|
"step": 680
|
|
},
|
|
{
|
|
"epoch": 170.0,
|
|
"eval_loss": 0.01307722833007574,
|
|
"eval_runtime": 0.2514,
|
|
"eval_samples_per_second": 493.221,
|
|
"eval_steps_per_second": 15.91,
|
|
"step": 680
|
|
},
|
|
{
|
|
"epoch": 171.0,
|
|
"eval_loss": 0.013027260079979897,
|
|
"eval_runtime": 0.244,
|
|
"eval_samples_per_second": 508.291,
|
|
"eval_steps_per_second": 16.396,
|
|
"step": 684
|
|
},
|
|
{
|
|
"epoch": 172.0,
|
|
"eval_loss": 0.01299793366342783,
|
|
"eval_runtime": 0.2446,
|
|
"eval_samples_per_second": 506.985,
|
|
"eval_steps_per_second": 16.354,
|
|
"step": 688
|
|
},
|
|
{
|
|
"epoch": 172.5,
|
|
"learning_rate": 2.6156127639784618e-06,
|
|
"loss": 0.0114,
|
|
"step": 690
|
|
},
|
|
{
|
|
"epoch": 173.0,
|
|
"eval_loss": 0.012975768186151981,
|
|
"eval_runtime": 0.235,
|
|
"eval_samples_per_second": 527.661,
|
|
"eval_steps_per_second": 17.021,
|
|
"step": 692
|
|
},
|
|
{
|
|
"epoch": 174.0,
|
|
"eval_loss": 0.012935076840221882,
|
|
"eval_runtime": 0.2413,
|
|
"eval_samples_per_second": 513.802,
|
|
"eval_steps_per_second": 16.574,
|
|
"step": 696
|
|
},
|
|
{
|
|
"epoch": 175.0,
|
|
"learning_rate": 2.183372119961499e-06,
|
|
"loss": 0.0114,
|
|
"step": 700
|
|
},
|
|
{
|
|
"epoch": 175.0,
|
|
"eval_loss": 0.012902160175144672,
|
|
"eval_runtime": 0.2344,
|
|
"eval_samples_per_second": 529.035,
|
|
"eval_steps_per_second": 17.066,
|
|
"step": 700
|
|
},
|
|
{
|
|
"epoch": 176.0,
|
|
"eval_loss": 0.012853474356234074,
|
|
"eval_runtime": 0.2359,
|
|
"eval_samples_per_second": 525.562,
|
|
"eval_steps_per_second": 16.954,
|
|
"step": 704
|
|
},
|
|
{
|
|
"epoch": 177.0,
|
|
"eval_loss": 0.012847784906625748,
|
|
"eval_runtime": 0.2402,
|
|
"eval_samples_per_second": 516.337,
|
|
"eval_steps_per_second": 16.656,
|
|
"step": 708
|
|
},
|
|
{
|
|
"epoch": 177.5,
|
|
"learning_rate": 1.7885225934020588e-06,
|
|
"loss": 0.0116,
|
|
"step": 710
|
|
},
|
|
{
|
|
"epoch": 178.0,
|
|
"eval_loss": 0.012830231338739395,
|
|
"eval_runtime": 0.254,
|
|
"eval_samples_per_second": 488.149,
|
|
"eval_steps_per_second": 15.747,
|
|
"step": 712
|
|
},
|
|
{
|
|
"epoch": 179.0,
|
|
"eval_loss": 0.012772643007338047,
|
|
"eval_runtime": 0.2357,
|
|
"eval_samples_per_second": 526.054,
|
|
"eval_steps_per_second": 16.969,
|
|
"step": 716
|
|
},
|
|
{
|
|
"epoch": 180.0,
|
|
"learning_rate": 1.4317112503391432e-06,
|
|
"loss": 0.011,
|
|
"step": 720
|
|
},
|
|
{
|
|
"epoch": 180.0,
|
|
"eval_loss": 0.012718794867396355,
|
|
"eval_runtime": 0.2414,
|
|
"eval_samples_per_second": 513.663,
|
|
"eval_steps_per_second": 16.57,
|
|
"step": 720
|
|
},
|
|
{
|
|
"epoch": 181.0,
|
|
"eval_loss": 0.012649140320718288,
|
|
"eval_runtime": 0.2405,
|
|
"eval_samples_per_second": 515.596,
|
|
"eval_steps_per_second": 16.632,
|
|
"step": 724
|
|
},
|
|
{
|
|
"epoch": 182.0,
|
|
"eval_loss": 0.012612031772732735,
|
|
"eval_runtime": 0.2432,
|
|
"eval_samples_per_second": 509.906,
|
|
"eval_steps_per_second": 16.449,
|
|
"step": 728
|
|
},
|
|
{
|
|
"epoch": 182.5,
|
|
"learning_rate": 1.1135228211241828e-06,
|
|
"loss": 0.0109,
|
|
"step": 730
|
|
},
|
|
{
|
|
"epoch": 183.0,
|
|
"eval_loss": 0.012596582062542439,
|
|
"eval_runtime": 0.2494,
|
|
"eval_samples_per_second": 497.26,
|
|
"eval_steps_per_second": 16.041,
|
|
"step": 732
|
|
},
|
|
{
|
|
"epoch": 184.0,
|
|
"eval_loss": 0.012583269737660885,
|
|
"eval_runtime": 0.2423,
|
|
"eval_samples_per_second": 511.721,
|
|
"eval_steps_per_second": 16.507,
|
|
"step": 736
|
|
},
|
|
{
|
|
"epoch": 185.0,
|
|
"learning_rate": 8.344787421847217e-07,
|
|
"loss": 0.0111,
|
|
"step": 740
|
|
},
|
|
{
|
|
"epoch": 185.0,
|
|
"eval_loss": 0.012561668641865253,
|
|
"eval_runtime": 0.2401,
|
|
"eval_samples_per_second": 516.548,
|
|
"eval_steps_per_second": 16.663,
|
|
"step": 740
|
|
},
|
|
{
|
|
"epoch": 186.0,
|
|
"eval_loss": 0.012579131871461868,
|
|
"eval_runtime": 0.2462,
|
|
"eval_samples_per_second": 503.612,
|
|
"eval_steps_per_second": 16.246,
|
|
"step": 744
|
|
},
|
|
{
|
|
"epoch": 187.0,
|
|
"eval_loss": 0.012605368159711361,
|
|
"eval_runtime": 0.2498,
|
|
"eval_samples_per_second": 496.418,
|
|
"eval_steps_per_second": 16.013,
|
|
"step": 748
|
|
},
|
|
{
|
|
"epoch": 187.5,
|
|
"learning_rate": 5.950363015120503e-07,
|
|
"loss": 0.0112,
|
|
"step": 750
|
|
},
|
|
{
|
|
"epoch": 188.0,
|
|
"eval_loss": 0.012568717822432518,
|
|
"eval_runtime": 0.2423,
|
|
"eval_samples_per_second": 511.759,
|
|
"eval_steps_per_second": 16.508,
|
|
"step": 752
|
|
},
|
|
{
|
|
"epoch": 189.0,
|
|
"eval_loss": 0.012545042671263218,
|
|
"eval_runtime": 0.2419,
|
|
"eval_samples_per_second": 512.52,
|
|
"eval_steps_per_second": 16.533,
|
|
"step": 756
|
|
},
|
|
{
|
|
"epoch": 190.0,
|
|
"learning_rate": 3.9558788927314407e-07,
|
|
"loss": 0.0108,
|
|
"step": 760
|
|
},
|
|
{
|
|
"epoch": 190.0,
|
|
"eval_loss": 0.012547217309474945,
|
|
"eval_runtime": 0.2425,
|
|
"eval_samples_per_second": 511.435,
|
|
"eval_steps_per_second": 16.498,
|
|
"step": 760
|
|
},
|
|
{
|
|
"epoch": 191.0,
|
|
"eval_loss": 0.012541635893285275,
|
|
"eval_runtime": 0.2428,
|
|
"eval_samples_per_second": 510.735,
|
|
"eval_steps_per_second": 16.475,
|
|
"step": 764
|
|
},
|
|
{
|
|
"epoch": 192.0,
|
|
"eval_loss": 0.012527569197118282,
|
|
"eval_runtime": 0.2498,
|
|
"eval_samples_per_second": 496.367,
|
|
"eval_steps_per_second": 16.012,
|
|
"step": 768
|
|
},
|
|
{
|
|
"epoch": 192.5,
|
|
"learning_rate": 2.3646035477491723e-07,
|
|
"loss": 0.0105,
|
|
"step": 770
|
|
},
|
|
{
|
|
"epoch": 193.0,
|
|
"eval_loss": 0.012520045973360538,
|
|
"eval_runtime": 0.2418,
|
|
"eval_samples_per_second": 512.815,
|
|
"eval_steps_per_second": 16.542,
|
|
"step": 772
|
|
},
|
|
{
|
|
"epoch": 194.0,
|
|
"eval_loss": 0.012517533265054226,
|
|
"eval_runtime": 0.2424,
|
|
"eval_samples_per_second": 511.589,
|
|
"eval_steps_per_second": 16.503,
|
|
"step": 776
|
|
},
|
|
{
|
|
"epoch": 195.0,
|
|
"learning_rate": 1.1791447083465134e-07,
|
|
"loss": 0.0106,
|
|
"step": 780
|
|
},
|
|
{
|
|
"epoch": 195.0,
|
|
"eval_loss": 0.012520862743258476,
|
|
"eval_runtime": 0.2491,
|
|
"eval_samples_per_second": 497.702,
|
|
"eval_steps_per_second": 16.055,
|
|
"step": 780
|
|
},
|
|
{
|
|
"epoch": 196.0,
|
|
"eval_loss": 0.012515625916421413,
|
|
"eval_runtime": 0.2418,
|
|
"eval_samples_per_second": 512.76,
|
|
"eval_steps_per_second": 16.541,
|
|
"step": 784
|
|
},
|
|
{
|
|
"epoch": 197.0,
|
|
"eval_loss": 0.012514699250459671,
|
|
"eval_runtime": 0.2439,
|
|
"eval_samples_per_second": 508.487,
|
|
"eval_steps_per_second": 16.403,
|
|
"step": 788
|
|
},
|
|
{
|
|
"epoch": 197.5,
|
|
"learning_rate": 4.01445064343281e-08,
|
|
"loss": 0.011,
|
|
"step": 790
|
|
},
|
|
{
|
|
"epoch": 198.0,
|
|
"eval_loss": 0.012506754137575626,
|
|
"eval_runtime": 0.2423,
|
|
"eval_samples_per_second": 511.845,
|
|
"eval_steps_per_second": 16.511,
|
|
"step": 792
|
|
},
|
|
{
|
|
"epoch": 199.0,
|
|
"eval_loss": 0.012521080672740936,
|
|
"eval_runtime": 0.2439,
|
|
"eval_samples_per_second": 508.345,
|
|
"eval_steps_per_second": 16.398,
|
|
"step": 796
|
|
},
|
|
{
|
|
"epoch": 200.0,
|
|
"learning_rate": 3.2779083591949478e-09,
|
|
"loss": 0.0111,
|
|
"step": 800
|
|
},
|
|
{
|
|
"epoch": 200.0,
|
|
"eval_loss": 0.012517580762505531,
|
|
"eval_runtime": 0.2414,
|
|
"eval_samples_per_second": 513.763,
|
|
"eval_steps_per_second": 16.573,
|
|
"step": 800
|
|
},
|
|
{
|
|
"epoch": 200.0,
|
|
"step": 800,
|
|
"total_flos": 4.3394125409969766e+17,
|
|
"train_loss": 0.44117489744909105,
|
|
"train_runtime": 692.9833,
|
|
"train_samples_per_second": 142.861,
|
|
"train_steps_per_second": 1.154
|
|
}
|
|
],
|
|
"max_steps": 800,
|
|
"num_train_epochs": 200,
|
|
"total_flos": 4.3394125409969766e+17,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|