Upload LoRA adapter (README written by author)

README.md

@@ -1,5 +1,5 @@
 ---
-base_model: Qwen/Qwen3-4B-Instruct-2507
+base_model: unsloth/Qwen3-4B-Instruct-2507
 datasets:
 - Tentoumaru/structured_data_with_cot_dataset_512_v2_nocot_whit_rules
 language:
@@ -13,10 +13,10 @@ tags:
 - structured-output
 ---
 
-＜qwen3-4b-dataset_512_v2_nocot_whit_rules＞
+＜qwen3-4b-dataset_512_v2_nocot_whit_rules2＞
 
 This repository provides a **LoRA adapter** fine-tuned from
-**Qwen/Qwen3-4B-Instruct-2507** using **QLoRA (4-bit, Unsloth)**.
+**unsloth/Qwen3-4B-Instruct-2507** using **QLoRA (4-bit, Unsloth)**.
 
 This repository contains **LoRA adapter weights only**.
 The base model must be loaded separately.
@@ -31,7 +31,7 @@ while intermediate reasoning (Chain-of-Thought) is masked.
 
 ## Training Configuration
 
-- Base model: Qwen/Qwen3-4B-Instruct-2507
+- Base model: unsloth/Qwen3-4B-Instruct-2507
 - Method: QLoRA (4-bit)
 - Max sequence length: 512
 - Epochs: 1
@@ -45,7 +45,7 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 from peft import PeftModel
 import torch
 
-base = "Qwen/Qwen3-4B-Instruct-2507"
+base = "unsloth/Qwen3-4B-Instruct-2507"
 adapter = "Tentoumaru/structured_data_with_cot_dataset_512_v2_nocot"
 
 tokenizer = AutoTokenizer.from_pretrained(base)

adapter_config.json

@@ -34,12 +34,12 @@
   "revision": null,
   "target_modules": [
     "up_proj",
-    "v_proj",
-    "q_proj",
-    "gate_proj",
-    "down_proj",
     "k_proj",
-    "o_proj"
+    "down_proj",
+    "gate_proj",
+    "o_proj",
+    "v_proj",
+    "q_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:b62f497b65be420e19e0d5cddf4661132ab8b41e39fafec09e676ef36b1647c9
+oid sha256:a3b1d6af081c9fedf7a86ee1fc9c8a255c5041e7a4afa6e5111817bce4d30fc3
 size 528550256