Model: Hyeongwon/P9-split1_only_answer_Qwen3-4B-Base_0402-01-1e-5 Source: Original Platform
9 lines
233 B
JSON
9 lines
233 B
JSON
{
|
|
"epoch": 6.0,
|
|
"total_flos": 515196244262912.0,
|
|
"train_loss": 0.37932722877818986,
|
|
"train_runtime": 69903.6938,
|
|
"train_samples": 41030,
|
|
"train_samples_per_second": 3.522,
|
|
"train_steps_per_second": 0.008
|
|
} |