Best model so far - eval_loss: 0.5369 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5499 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5694 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5745 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.5888 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6107 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.6416 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 1.1570 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.1629 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.1717 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 1.1447 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.1520 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.1846 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.7053 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7088 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7250 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7392 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.6992 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7010 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7182 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7334 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 1.0370 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.0426 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.0658 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.1013 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.2204 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.9854 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.9912 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.9927 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.9972 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.0150 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.0342 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.0621 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.1015 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.2240 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 1.4867 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.4934 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.4983 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5021 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5179 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 1.5485 -------------------------------------------------------- (Trained with Unsloth)

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

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