Best model so far - eval_loss: 1.2499 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.2524 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.2551 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.2591 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.2669 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.2751 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.2918 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.2999 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.3202 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.3412 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.3648 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.4142 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.4661 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5701 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7058 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.9490 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 2.2399 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 1.9811 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 2.2973 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.1130 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.1220 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.1283 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.1301 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.1671 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.1745 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.2053 -------------------------------------------------------- (Trained with Unsloth)

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

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

Best model so far - eval_loss: 0.0845 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0848 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0848 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0852 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0891 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0919 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0940 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.1065 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.1191 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.1315 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.1414 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.1444 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.1684 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.1732 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.1875 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.2021 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 7.4271 --------------------------------------------------------

Best model so far - eval_loss: 7.6829 --------------------------------------------------------

Initial model state ------------------------------------------------------------------------