Upload LoRA adapter (README written by author)

README.md

@@ -33,9 +33,9 @@ while intermediate reasoning (Chain-of-Thought) is masked.
 
 - Base model: Qwen/Qwen3-4B-Instruct-2507
 - Method: QLoRA (4-bit)
-- Max sequence length: 512
-- Epochs: 2
-- Learning rate: 2e-06
+- Max sequence length: 1024
+- Epochs: 1
+- Learning rate: 1e-06
 - LoRA: r=64, alpha=128
 
 ## Usage

adapter_config.json

@@ -33,13 +33,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
     "down_proj",
-    "gate_proj",
-    "o_proj",
+    "q_proj",
     "v_proj",
+    "k_proj",
+    "o_proj",
     "up_proj",
-    "k_proj"
+    "gate_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ac8ccce16f52ddea4599ac17ab5ba04d92aa443c93c483806949bcc5983dbc51
+oid sha256:12f1081ce63a01aae40500966d28b5575d15330263feb4e10f7f98e73caff8b7
 size 528550256