Best model so far - eval_loss: 1.5780 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.6308 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7360 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8675 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 1.5125 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5463 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5495 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5737 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.6340 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7336 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8684 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 1.4809 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.4857 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.4910 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.4942 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5150 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5545 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5573 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5724 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.6374 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7351 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8683 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 1.5465 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5828 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.6319 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7498 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8668 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 1.5078 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5202 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5478 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5688 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.6319 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7389 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8675 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 1.5867 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.6326 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.7430 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.8676 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 1.1730 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.2405 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 1.0752 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.0900 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.0971 -------------------------------------------------------- (Trained with Unsloth)