146 lines
3.9 KiB
JSON
146 lines
3.9 KiB
JSON
|
[
|
||
|
|
{
|
||
|
|
"loss": 1.8184,
|
||
|
|
"grad_norm": 0.774284303188324,
|
||
|
|
"learning_rate": 8.923190911336132e-05,
|
||
|
|
"epoch": 0.2218976306523778,
|
||
|
|
"step": 451
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"eval_loss": 1.749639630317688,
|
||
|
|
"eval_runtime": 1083.3199,
|
||
|
|
"eval_samples_per_second": 26.684,
|
||
|
|
"eval_steps_per_second": 6.671,
|
||
|
|
"epoch": 0.2218976306523778,
|
||
|
|
"step": 451
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"loss": 1.7064,
|
||
|
|
"grad_norm": 0.7226008772850037,
|
||
|
|
"learning_rate": 7.809335638429242e-05,
|
||
|
|
"epoch": 0.4437952613047556,
|
||
|
|
"step": 902
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"eval_loss": 1.6820720434188843,
|
||
|
|
"eval_runtime": 1087.8354,
|
||
|
|
"eval_samples_per_second": 26.573,
|
||
|
|
"eval_steps_per_second": 6.643,
|
||
|
|
"epoch": 0.4437952613047556,
|
||
|
|
"step": 902
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"loss": 1.6623,
|
||
|
|
"grad_norm": 0.7210889458656311,
|
||
|
|
"learning_rate": 6.695480365522352e-05,
|
||
|
|
"epoch": 0.6656928919571334,
|
||
|
|
"step": 1353
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"eval_loss": 1.6484241485595703,
|
||
|
|
"eval_runtime": 1084.0066,
|
||
|
|
"eval_samples_per_second": 26.667,
|
||
|
|
"eval_steps_per_second": 6.667,
|
||
|
|
"epoch": 0.6656928919571334,
|
||
|
|
"step": 1353
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"loss": 1.6362,
|
||
|
|
"grad_norm": 0.7425829172134399,
|
||
|
|
"learning_rate": 5.581625092615461e-05,
|
||
|
|
"epoch": 0.8875905226095112,
|
||
|
|
"step": 1804
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"eval_loss": 1.6261749267578125,
|
||
|
|
"eval_runtime": 1085.6726,
|
||
|
|
"eval_samples_per_second": 26.626,
|
||
|
|
"eval_steps_per_second": 6.657,
|
||
|
|
"epoch": 0.8875905226095112,
|
||
|
|
"step": 1804
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"loss": 1.5914,
|
||
|
|
"grad_norm": 0.7476176023483276,
|
||
|
|
"learning_rate": 4.4677698197085704e-05,
|
||
|
|
"epoch": 1.109488153261889,
|
||
|
|
"step": 2255
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"eval_loss": 1.6124544143676758,
|
||
|
|
"eval_runtime": 1084.6953,
|
||
|
|
"eval_samples_per_second": 26.65,
|
||
|
|
"eval_steps_per_second": 6.663,
|
||
|
|
"epoch": 1.109488153261889,
|
||
|
|
"step": 2255
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"loss": 1.5557,
|
||
|
|
"grad_norm": 0.7473255395889282,
|
||
|
|
"learning_rate": 3.3539145468016795e-05,
|
||
|
|
"epoch": 1.3313857839142669,
|
||
|
|
"step": 2706
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"eval_loss": 1.6011990308761597,
|
||
|
|
"eval_runtime": 1084.5737,
|
||
|
|
"eval_samples_per_second": 26.653,
|
||
|
|
"eval_steps_per_second": 6.663,
|
||
|
|
"epoch": 1.3313857839142669,
|
||
|
|
"step": 2706
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"loss": 1.5468,
|
||
|
|
"grad_norm": 0.750347375869751,
|
||
|
|
"learning_rate": 2.240059273894789e-05,
|
||
|
|
"epoch": 1.5532834145666445,
|
||
|
|
"step": 3157
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"eval_loss": 1.592125415802002,
|
||
|
|
"eval_runtime": 1086.1382,
|
||
|
|
"eval_samples_per_second": 26.614,
|
||
|
|
"eval_steps_per_second": 6.654,
|
||
|
|
"epoch": 1.5532834145666445,
|
||
|
|
"step": 3157
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"loss": 1.539,
|
||
|
|
"grad_norm": 0.7452530860900879,
|
||
|
|
"learning_rate": 1.1262040009878982e-05,
|
||
|
|
"epoch": 1.7751810452190224,
|
||
|
|
"step": 3608
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"eval_loss": 1.5855711698532104,
|
||
|
|
"eval_runtime": 1084.064,
|
||
|
|
"eval_samples_per_second": 26.665,
|
||
|
|
"eval_steps_per_second": 6.667,
|
||
|
|
"epoch": 1.7751810452190224,
|
||
|
|
"step": 3608
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"loss": 1.5373,
|
||
|
|
"grad_norm": 0.7762174606323242,
|
||
|
|
"learning_rate": 1.2348728081007656e-07,
|
||
|
|
"epoch": 1.9970786758714003,
|
||
|
|
"step": 4059
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"eval_loss": 1.5827687978744507,
|
||
|
|
"eval_runtime": 1086.4478,
|
||
|
|
"eval_samples_per_second": 26.607,
|
||
|
|
"eval_steps_per_second": 6.652,
|
||
|
|
"epoch": 1.9970786758714003,
|
||
|
|
"step": 4059
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"train_runtime": 69830.4547,
|
||
|
|
"train_samples_per_second": 7.451,
|
||
|
|
"train_steps_per_second": 0.058,
|
||
|
|
"total_flos": 9.47865607059515e+18,
|
||
|
|
"train_loss": 1.5162640926171476,
|
||
|
|
"epoch": 1.9995387382954841,
|
||
|
|
"step": 4064
|
||
|
|
}
|
||
|
|
]
|