Upload LoRA adapter (README written by author)

README.md

@@ -1,7 +1,7 @@
 ---
 base_model: Qwen/Qwen3-4B-Instruct-2507
 datasets:
-- u-10bei/structured_data_with_cot_dataset_512_v5
+- u-10bei/structured_data_with_cot_dataset_512_v2
 language:
 - en
 license: apache-2.0
@@ -13,7 +13,7 @@ tags:
 - structured-output
 ---
 
-＜【課題】環境設定変更１+データ変更１＞
+＜【課題】qwen3-4b-structured-output-lora＞
 
 This repository provides a **LoRA adapter** fine-tuned from
 **Qwen/Qwen3-4B-Instruct-2507** using **QLoRA (4-bit, Unsloth)**.
@@ -34,9 +34,9 @@ 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-05
-- LoRA: r=64, alpha=128
+- Epochs: 3
+- Learning rate: 1e-05
+- LoRA: r=128, alpha=128
 
 ## Usage
 
@@ -59,7 +59,7 @@ model = PeftModel.from_pretrained(model, adapter)
 
 ## Sources & Terms (IMPORTANT)
 
-Training data: u-10bei/structured_data_with_cot_dataset_512_v5
+Training data: u-10bei/structured_data_with_cot_dataset_512_v2
 
 Dataset License: MIT License. This dataset is used and distributed under the terms of the MIT License.
 Compliance: Users must comply with the MIT license (including copyright notice) and the base model's original terms of use.

adapter_config.json

@@ -29,17 +29,17 @@
   "peft_type": "LORA",
   "peft_version": "0.18.1",
   "qalora_group_size": 16,
-  "r": 64,
+  "r": 128,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
-    "down_proj",
     "q_proj",
-    "v_proj",
-    "gate_proj",
+    "up_proj",
     "o_proj",
-    "up_proj"
+    "gate_proj",
+    "k_proj",
+    "down_proj",
+    "v_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:ac277601a5f080920bfaf129cef05b409b27a575acf8b2ae906a999a8eaa8588
-size 528550256
+oid sha256:de7f6ccd5ab2d26d948bce55548b991365b117b79796ff7ac66856cca801a8d5
+size 1057033224