qwen-2.5-3b-r1-countdown / train_results.json
mgaimm's picture
Model save
b315b7b verified
raw
history blame contribute delete
204 Bytes
{
"total_flos": 0.0,
"train_loss": 0.00020930594997387746,
"train_runtime": 20107.4364,
"train_samples": 45000,
"train_samples_per_second": 0.537,
"train_steps_per_second": 0.022
}