Upload 8 files

README.md

@@ -34,9 +34,11 @@ while intermediate reasoning (Chain-of-Thought) is masked.
 - Base model: Qwen/Qwen3-4B-Instruct-2507
 - Method: QLoRA (4-bit)
 - Max sequence length: 2048
-- Epochs: 2
-- Learning rate: 2e-05
-- LoRA: r=32, alpha=64
+- Epochs: 3
+- Learning rate: 1e-06
+- Weight decay: 0.03
+- Warmup ratio: 0.03
+- LoRA: r=64, alpha=128, dropout=0.02
 
 ## Usage
 
@@ -46,7 +48,7 @@ from peft import PeftModel
 import torch
 
 base = "Qwen/Qwen3-4B-Instruct-2507"
-adapter = "your_id/your-repo"
+adapter = "MF0619/LLM2025A"
 
 tokenizer = AutoTokenizer.from_pretrained(base)
 model = AutoModelForCausalLM.from_pretrained(

adapter_config.json

@@ -20,26 +20,26 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 64,
+  "lora_alpha": 128,
   "lora_bias": false,
-  "lora_dropout": 0.05,
+  "lora_dropout": 0.02,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
   "peft_version": "0.18.1",
   "qalora_group_size": 16,
-  "r": 32,
+  "r": 64,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
+    "up_proj",
     "o_proj",
-    "down_proj",
-    "gate_proj",
     "q_proj",
-    "up_proj",
-    "k_proj"
+    "k_proj",
+    "down_proj",
+    "v_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:b691675d975607c342b28d8000efb8351664544a04ed80643eba37ea51c891cd
-size 264308896
+oid sha256:ce91beabf984431132ec016efe9ad787a80f600beba8c918542842ed3204eeec
+size 528550256