Best model so far - eval_loss: 0.5548 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5600 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5656 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5686 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5714 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5720 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5724 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5739 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5746 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5775 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5810 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5883 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5951 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6136 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6329 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6591 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6704 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6916 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7039 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7370 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7651 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7965 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8102 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8113 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8240 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8487 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8692 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.9155 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.9482 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.0101 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.0874 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.1773 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.2992 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.4505 -------------------------------------------------------- (Trained with Unsloth)

Initial model state ------------------------------------------------------------------------ (Trained with Unsloth)

initial commit