mkopecki
/

chess-lora-adapter-fp-llama-3.1-8b

Generated from Trainer

Model card Files Files and versions Community

mkopecki commited on Aug 8

Commit

23eb1c2

•

1 Parent(s): 3934b0d

End of training

Files changed (1) hide show

trainer_state.json +3 -3

trainer_state.json CHANGED Viewed

@@ -12,9 +12,9 @@
       "epoch": 3.0,
       "step": 408,
       "total_flos": 2.2972323251552256e+16,
-      "train_loss": 0.48282754187490423,
-      "train_runtime": 596.2942,
-      "train_samples_per_second": 2.727,
       "train_steps_per_second": 0.684
     }
   ],

       "epoch": 3.0,
       "step": 408,
       "total_flos": 2.2972323251552256e+16,
+      "train_loss": 0.4828735800350414,
+      "train_runtime": 596.4455,
+      "train_samples_per_second": 2.726,
       "train_steps_per_second": 0.684
     }
   ],