Qwen-2.5-7B-Simple-RL-v2 / train_results.json
aidando73's picture
.
c381217
{
"total_flos": 0.0,
"train_loss": 0.06606895855531611,
"train_runtime": 9129.9094,
"train_samples": 7500,
"train_samples_per_second": 0.821,
"train_steps_per_second": 0.006
}