Qwen2.5-Math-7B-Instruct-eos / all_results.json
flyingbugs's picture
Model save
85c6a7c verified
raw
history blame contribute delete
219 Bytes
{
"total_flos": 2.755855970469937e+18,
"train_loss": 0.4114767361979708,
"train_runtime": 186765.2867,
"train_samples": 93733,
"train_samples_per_second": 0.276,
"train_steps_per_second": 0.017
}