Best model so far - eval_loss: 1.7660 --------------------------------------------------------

Best model so far - eval_loss: 0.5032 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8191 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8144 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8121 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8201 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8243 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6778 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6815 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6845 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6881 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6974 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7005 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7101 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7123 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7298 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7251 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7650 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7469 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7734 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7683 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8364 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8111 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8325 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8105 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 6.7761 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 7.1735 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 7.5552 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 6.7628 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 6.7710 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 7.5888 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 7.6503 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 6.7494 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 6.7853 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 7.5588 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 6.7300 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 6.7615 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 9.8748 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 7.5755 -------------------------------------------------------- (Trained with Unsloth)

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

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

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

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

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

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

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

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

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

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

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