Best model so far - eval_loss: 0.0189 --------------------------------------------------------

Best model so far - eval_loss: 0.0199 --------------------------------------------------------

Best model so far - eval_loss: 0.0217 --------------------------------------------------------

Initial model state ------------------------------------------------------------------------

Best model so far - eval_loss: 0.0067 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.0073 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 2.2034 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 2.1930 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 2.3682 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 2.5022 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 2.6305 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 2.7126 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.0021 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0021 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0022 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0023 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0029 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0031 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0031 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0034 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0043 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.0063 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.7968 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8011 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8089 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.8160 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8196 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8251 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.8159 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.8013 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.8014 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.7950 -------------------------------------------------------- (Trained with Unsloth)

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

Best model so far - eval_loss: 0.7779 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7810 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7871 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 0.7992 -------------------------------------------------------- (Trained with Unsloth)