Qwen2.5-1.5B-Instruct_Long_CoT / all_results.json
TaiGary's picture
Add files using upload-large-folder tool
2050225 verified
raw
history blame contribute delete
343 Bytes
{
"epoch": 2.0,
"eval_loss": 0.36304256319999695,
"eval_runtime": 2.0837,
"eval_samples_per_second": 25.915,
"eval_steps_per_second": 6.719,
"total_flos": 21199392964608.0,
"train_loss": 0.3376133635327845,
"train_runtime": 1178.6141,
"train_samples_per_second": 9.043,
"train_steps_per_second": 0.567
}