ahmetbagci
hello
d3d65a7
raw
history blame
3.97 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"global_step": 11580,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.13,
"learning_rate": 4.784110535405873e-05,
"loss": 4.4347,
"step": 500
},
{
"epoch": 0.26,
"learning_rate": 4.5682210708117445e-05,
"loss": 3.0223,
"step": 1000
},
{
"epoch": 0.39,
"learning_rate": 4.352331606217617e-05,
"loss": 2.5837,
"step": 1500
},
{
"epoch": 0.52,
"learning_rate": 4.136442141623489e-05,
"loss": 2.3595,
"step": 2000
},
{
"epoch": 0.65,
"learning_rate": 3.920552677029361e-05,
"loss": 2.225,
"step": 2500
},
{
"epoch": 0.78,
"learning_rate": 3.704663212435233e-05,
"loss": 2.0827,
"step": 3000
},
{
"epoch": 0.91,
"learning_rate": 3.4887737478411055e-05,
"loss": 1.9716,
"step": 3500
},
{
"epoch": 1.0,
"eval_loss": 1.879624843597412,
"eval_rouge2_fmeasure": 0.2999,
"eval_rouge2_precision": 0.326,
"eval_rouge2_recall": 0.2944,
"eval_runtime": 57.7084,
"eval_samples_per_second": 20.794,
"step": 3860
},
{
"epoch": 1.04,
"learning_rate": 3.272884283246977e-05,
"loss": 1.8231,
"step": 4000
},
{
"epoch": 1.17,
"learning_rate": 3.05699481865285e-05,
"loss": 1.5489,
"step": 4500
},
{
"epoch": 1.3,
"learning_rate": 2.841105354058722e-05,
"loss": 1.5264,
"step": 5000
},
{
"epoch": 1.42,
"learning_rate": 2.6252158894645944e-05,
"loss": 1.5027,
"step": 5500
},
{
"epoch": 1.55,
"learning_rate": 2.4093264248704665e-05,
"loss": 1.4792,
"step": 6000
},
{
"epoch": 1.68,
"learning_rate": 2.1934369602763387e-05,
"loss": 1.4515,
"step": 6500
},
{
"epoch": 1.81,
"learning_rate": 1.9775474956822108e-05,
"loss": 1.4273,
"step": 7000
},
{
"epoch": 1.94,
"learning_rate": 1.761658031088083e-05,
"loss": 1.418,
"step": 7500
},
{
"epoch": 2.0,
"eval_loss": 1.6983058452606201,
"eval_rouge2_fmeasure": 0.3188,
"eval_rouge2_precision": 0.3398,
"eval_rouge2_recall": 0.3172,
"eval_runtime": 59.4641,
"eval_samples_per_second": 20.18,
"step": 7720
},
{
"epoch": 2.07,
"learning_rate": 1.5457685664939554e-05,
"loss": 1.2305,
"step": 8000
},
{
"epoch": 2.2,
"learning_rate": 1.3298791018998274e-05,
"loss": 1.0815,
"step": 8500
},
{
"epoch": 2.33,
"learning_rate": 1.1139896373056995e-05,
"loss": 1.1018,
"step": 9000
},
{
"epoch": 2.46,
"learning_rate": 8.981001727115718e-06,
"loss": 1.0829,
"step": 9500
},
{
"epoch": 2.59,
"learning_rate": 6.822107081174439e-06,
"loss": 1.0609,
"step": 10000
},
{
"epoch": 2.72,
"learning_rate": 4.663212435233161e-06,
"loss": 1.0557,
"step": 10500
},
{
"epoch": 2.85,
"learning_rate": 2.504317789291883e-06,
"loss": 1.0546,
"step": 11000
},
{
"epoch": 2.98,
"learning_rate": 3.454231433506045e-07,
"loss": 1.0538,
"step": 11500
},
{
"epoch": 3.0,
"eval_loss": 1.6779444217681885,
"eval_rouge2_fmeasure": 0.3222,
"eval_rouge2_precision": 0.3433,
"eval_rouge2_recall": 0.3198,
"eval_runtime": 60.5062,
"eval_samples_per_second": 19.833,
"step": 11580
}
],
"max_steps": 11580,
"num_train_epochs": 3,
"total_flos": 22201145347276800,
"trial_name": null,
"trial_params": null
}