Best model so far - eval_loss: 1.7638 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7639 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7641 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7647 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7648 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7656 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7663 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7669 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7682 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7696 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7711 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7729 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7745 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7767 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7775 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7790 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7812 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7814 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7818 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7836 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7851 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7881 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7900 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7933 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7965 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7995 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8015 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8035 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8105 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8154 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8206 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8215 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8225 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8254 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8304 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8377 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8455 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8565 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8708 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8860 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.9066 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.9338 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.9785 -------------------------------------------------------- (Trained with Unsloth)

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

initial commit