Model save

README.md

@@ -42,7 +42,7 @@ The following hyperparameters were used during training:
 - optimizer: Use OptimizerNames.ADAMW_HF with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 2
-- num_epochs: 1
+- num_epochs: 3
 - mixed_precision_training: Native AMP
 
 ### Training results

adapter_config.json

@@ -26,13 +26,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "gate_proj",
-    "k_proj",
-    "v_proj",
-    "up_proj",
     "q_proj",
+    "down_proj",
     "o_proj",
-    "down_proj"
+    "up_proj",
+    "k_proj",
+    "v_proj",
+    "gate_proj"
   ],
   "task_type": null,
   "use_dora": true,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d41d25df148987ef0c59580ed5c8ab70e453887f7380e65aa1c8df4728aa658
+oid sha256:6f4f8c67ae8f013a461c79314ff34f0bb5c1d9854b596c000a41300e7a313b79
 size 125866776

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:16fedcbe9fda65bb0f96d12da2ee23d15b2943987da9d0c9eee73191d6985587
+oid sha256:84e999e3244a488ff978b7bbc08d32a5624d64be19592b2aeafd106d86ce3067
 size 5304