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:c7d876d05da67f78e8e050d42e21e36f9295b984da2d5cc7175ed912ed48c683
-size 109086416
+oid sha256:9a5edf63af6fd01c3b488de0ad5b77b3a645fc523c4af8ba573b98a4887f03ea
+size 872450448

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3a0566d5a3feec6ced19b143047c018d20c5d2047003721454b7593edeab92bb
+oid sha256:55d9dc81dae7c71378881d52f3b86f755294f52a09db1c17cd0cc99a144175a2
 size 5048