Best model so far - eval_loss: 0.6366 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6482 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6578 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6643 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6865 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6979 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7174 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7553 -------------------------------------------------------- (Trained with Unsloth)

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

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

initial commit