Files
llama-3-8b-base-r-dpo-ultra…/eval_results.json
ModelHub XC 53d019704e 初始化项目,由ModelHub XC社区提供模型
Model: jackf857/llama-3-8b-base-r-dpo-ultrafeedback-4xH200-batch-128-rerun-2-runpod
Source: Original Platform
2026-05-09 22:20:05 +08:00

18 lines
698 B
JSON

{
"epoch": 0.9989528795811519,
"eval_logits/chosen": -0.34589293599128723,
"eval_logits/rejected": -0.3294574022293091,
"eval_logps/chosen": -2883.560791015625,
"eval_logps/ref_chosen": -288.6414794921875,
"eval_logps/ref_rejected": -265.96160888671875,
"eval_logps/rejected": -2528.9130859375,
"eval_loss": 0.6624810695648193,
"eval_r_dpo/chosen_len": 286.97601318359375,
"eval_r_dpo/length_delta": 40.88800048828125,
"eval_r_dpo/regularization_term": 4.088800430297852,
"eval_r_dpo/rejected_len": 246.08799743652344,
"eval_runtime": 196.5258,
"eval_samples": 2000,
"eval_samples_per_second": 10.177,
"eval_steps_per_second": 1.272
}