End of training

README.md

@@ -35,7 +35,7 @@ More information needed
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
 - train_batch_size: 3
-- eval_batch_size: 1
+- eval_batch_size: 3
 - seed: 42
 - gradient_accumulation_steps: 4
 - total_train_batch_size: 12

adapter_config.json

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

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d6783a789797fd7c6fbd35e7959c62ac136e598bd50d6654ee0fd05f372919b6
+oid sha256:ee1a2a4874006c104e9907eef517d568ebdca52d1f83e8b4e4d7507a2cdab90c
 size 19210776

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4742df7145dc76e19bd703cd57ff56344d836d2de82033e630767072b3db5d5d
+oid sha256:2b8141b2dc1ce1a784dd36bace65c11d72880a53816eff473c61fef26b636667
 size 5304