Best model so far - eval_loss: 7.2202 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 7.2330 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 7.2846 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 7.3423 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 7.2146 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 7.2361 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 7.2983 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 7.3679 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.6924 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6929 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6936 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6948 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6959 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6978 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7000 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7023 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7037 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7070 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7108 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7116 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7120 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7136 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7164 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7200 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7252 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7310 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.7443 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7533 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7626 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7798 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.7794 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.7397 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7454 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7536 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7625 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7808 -------------------------------------------------------- (Trained with Unsloth)

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

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

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

Best model so far - eval_loss: 0.7497 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7615 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7803 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.7580 -------------------------------------------------------- (Trained with Unsloth)

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