Files
deepseek-math-7B-finetuned/all_results.json

12 lines
348 B
JSON
Raw Normal View History

{
"epoch": 1.0,
"eval_loss": 0.05969985947012901,
"eval_runtime": 508.637,
"eval_samples_per_second": 11.474,
"eval_steps_per_second": 0.718,
"total_flos": 4644350000627712.0,
"train_loss": 0.08661497140495507,
"train_runtime": 70638.1483,
"train_samples_per_second": 2.677,
"train_steps_per_second": 0.084
}