Training in progress, epoch 1

adapter_config.json

@@ -13,7 +13,7 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 32,
+  "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"
+    "v_proj",
+    "o_proj",
+    "k_proj"
   ],
   "task_type": null,
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae48ce0ffd48fe1ec15240e99126770dd84cddaa4587d2ee72044d64f25dec8a
-size 109086416
+oid sha256:4e99395bd50aa2dfe05d90e1a1eafd90b7923728b9991b856996b926a80a2730
+size 872450448

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:30c68b632892d2e61672f66a7d6888440c5d6d681000dcf364633ff9907b3b91
+oid sha256:c3749b84915a3d0049c6dc404b6aae1b82d22fce5dcd66857f6e5b61282f510f
 size 5048