Model: jackf857/llama-3-8b-base-r-dpo-ultrafeedback-4xh200-batch-128-20260428-035521 Source: Original Platform
24 lines
880 B
JSON
24 lines
880 B
JSON
{
|
|
"epoch": 0.9989528795811519,
|
|
"eval_logits/chosen": -0.8675644993782043,
|
|
"eval_logits/rejected": -0.8504053950309753,
|
|
"eval_logps/chosen": -423.36651611328125,
|
|
"eval_logps/ref_chosen": -288.6414794921875,
|
|
"eval_logps/ref_rejected": -265.96160888671875,
|
|
"eval_logps/rejected": -462.2294616699219,
|
|
"eval_loss": 0.5316001772880554,
|
|
"eval_r_dpo/chosen_len": 286.97601318359375,
|
|
"eval_r_dpo/length_delta": 40.88800048828125,
|
|
"eval_r_dpo/regularization_term": 0.0,
|
|
"eval_r_dpo/rejected_len": 246.08799743652344,
|
|
"eval_runtime": 78.0724,
|
|
"eval_samples": 2000,
|
|
"eval_samples_per_second": 25.617,
|
|
"eval_steps_per_second": 3.202,
|
|
"total_flos": 0.0,
|
|
"train_loss": 4.583878276233153,
|
|
"train_runtime": 6810.0393,
|
|
"train_samples": 61135,
|
|
"train_samples_per_second": 8.977,
|
|
"train_steps_per_second": 0.07
|
|
} |