Training in progress, epoch 1

adapter_config.json

@@ -13,7 +13,7 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 64,
+  "lora_alpha": 256,
   "lora_dropout": 0.1,
   "megatron_config": null,
   "megatron_core": "megatron.core",
@@ -21,14 +21,14 @@
     "ln_f"
   ],
   "peft_type": "LORA",
-  "r": 32,
+  "r": 256,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
+    "k_proj",
     "q_proj",
-    "o_proj",
-    "k_proj"
+    "v_proj",
+    "o_proj"
   ],
   "task_type": null,
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a9d53210da30bc37bbefef065d4438a29bbb06f9211c7f8d556004e4b37f7288
-size 109086416
+oid sha256:7710c623843afc6f41c701a2b66dd1aa878f8f361a96b8a95efc0c459e97aa90
+size 872450448

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0881612d889f7c5b56ccfbdef24163696de100dc0091161e767614c175b235ca
+oid sha256:950610cdccb1edcc569bf338b6ddba41696e3d23a780218199d2310106cc4c00
 size 5048