CoolFace
Modelpublic

Logic12/DeepSeek-R1-7B-RedMind

sourceHugging Faceupdated 1y agoView on Hugging Face
0likes18downloads
train_results.json8 linesDownload Raw Back to root
1{2    "epoch": 5.0,3    "total_flos": 1.565023977603072e+16,4    "train_loss": 3.387373956044515,5    "train_runtime": 55.9294,6    "train_samples_per_second": 3.218,7    "train_steps_per_second": 0.2688}