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

Best model so far - eval_loss: 0.8192 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8317 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.7994 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7995 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.7857 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.8412 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8419 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8422 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8433 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8445 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8446 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8450 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8451 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8453 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8456 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8460 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8470 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8480 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8492 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8509 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8536 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8568 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8607 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8662 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8753 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8902 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.9216 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.5079 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5127 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5352 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5702 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6202 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6678 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7083 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7195 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 1.4737 -------------------------------------------------------- (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)

Best model so far - eval_loss: 0.4492 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.4641 -------------------------------------------------------- (Trained with Unsloth)