Upload LoRA adapter (README written by author)

README.md

@@ -36,7 +36,7 @@ while intermediate reasoning (Chain-of-Thought) is masked.
 - Max sequence length: 512
 - Epochs: 1
 - Learning rate: 1e-06
-- LoRA: r=64, alpha=128
+- LoRA: r=128, alpha=256
 
 ## Usage
 

adapter_config.json

@@ -20,7 +20,7 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 128,
+  "lora_alpha": 256,
   "lora_bias": false,
   "lora_dropout": 0.0,
   "megatron_config": null,
@@ -29,16 +29,16 @@
   "peft_type": "LORA",
   "peft_version": "0.18.1",
   "qalora_group_size": 16,
-  "r": 64,
+  "r": 128,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "gate_proj",
-    "v_proj",
-    "o_proj",
     "k_proj",
-    "down_proj",
+    "v_proj",
     "up_proj",
+    "down_proj",
+    "gate_proj",
+    "o_proj",
     "q_proj"
   ],
   "target_parameters": null,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:402ee459e4b160f08540b1d63a72dd0c472b54f974b35ea445d4a39ad06bec0b
-size 528550256
+oid sha256:1bac630dee8a66a1498d1a0c93272d055214ec8bdef162e351f22865ca5c6beb
+size 1057033224