Qwen-2.5-1.5B-Simple-RL-ga1 / all_results.json
Bradley's picture
Model save
0411cd2 verified
raw
history blame contribute delete
185 Bytes
{
"total_flos": 0.0,
"train_loss": 0.0,
"train_runtime": 1.7067,
"train_samples": 7500,
"train_samples_per_second": 4394.419,
"train_steps_per_second": 274.798
}