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 sha256:8127710ca95424e832e55704e4f7cee4d55a8a08ea447f710d210f5b3d3a826a
-size 109086416
+oid sha256:e5707eb2fa80fbeb6fcfe4fd76036c4036f923607f3751f7ab4d9c3ecf77725f
+size 872450448

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:77fdc9e2dba7cc02be65ae72881827fffdb3ac96f59ea05e7beb1826abcec3e3
+oid sha256:4049b9044a107df3911a2f47a93d96769630f3df3f1947f819b9b088dd3e3145
 size 5048