Training in progress, step 500

adapter_config.json

@@ -12,9 +12,9 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 128,
+  "lora_alpha": 64,
   "lora_bias": false,
-  "lora_dropout": 0.05,
+  "lora_dropout": 0.1,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
@@ -23,8 +23,8 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "qkv_proj"
+    "qkv_proj",
+    "o_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:680b61d7d897dc80a7b437d91066a57d1320e450a9389ba4b42ed22478034ca0
+oid sha256:4f52e2a2426b35ac9fa72680206bb94172f3b17fc12212ed9f09dd02373bfd7e
 size 943740600

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:425ba363d17319200beede7d472f921f2c78140502b2bf4a8e16ce51c94e0683
+oid sha256:f5faddc77408de9910d9c3e2f2ab1bbbd55ee43bf9cd4b27643d8adfd9f88d01
 size 5688