Training in progress, epoch 0

adapter_config.json

@@ -16,17 +16,17 @@
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 8,
+  "r": 4,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "gate_proj",
     "k_proj",
-    "q_proj",
+    "gate_proj",
+    "up_proj",
+    "v_proj",
     "o_proj",
     "down_proj",
-    "up_proj",
-    "v_proj"
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4250c02b4e7964a09d852ae901f53b35dfe408cde641039288d781f2cd8e2872
-size 19644912
+oid sha256:3f4907f0d8c87ff635f70f8ab6e869b5437dc6dbc6e289ad522e488cc7392620
+size 9838816

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:80048c5301a3f906cfcabaa1c46999ead76329777a58af2993a2f3e0fba9ec2b
+oid sha256:8c5d462de33cd5663080eae0ddb08967a0acabc1341ac4beee0ec69970e6ebb7
 size 5240