Best model so far - eval_loss: 10.3117 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3118 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3122 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3123 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3123 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3124 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3124 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3125 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3127 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3129 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3132 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3136 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3142 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3148 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3159 -------------------------------------------------------- (Trained with Unsloth)

Best model so far - eval_loss: 10.3175 -------------------------------------------------------- (Trained with Unsloth)

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

initial commit