GRPO LoRA (sft=1000, grpo=200)

adapter_config.json

@@ -7,7 +7,7 @@
     "parent_library": "transformers.models.qwen2.modeling_qwen2",
     "unsloth_fixed": true
   },
-  "base_model_name_or_path": "unsloth/qwen2.5-3b-instruct-unsloth-bnb-4bit",
+  "base_model_name_or_path": "unsloth/qwen2.5-coder-7b-instruct-bnb-4bit",
   "bias": "none",
   "corda_config": null,
   "ensure_weight_tying": false,
@@ -34,13 +34,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "down_proj",
-    "k_proj",
-    "o_proj",
     "up_proj",
+    "k_proj",
     "v_proj",
-    "gate_proj",
-    "q_proj"
+    "down_proj",
+    "q_proj",
+    "o_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 sha256:55e6151b16e8af4a8aaf0b7cfdac763ffd8825c85397ece9e1b582f64e6b5fba
-size 119801528
+oid sha256:b0b4c9947fcc438880784fc8fe35481f75c42270a97efedac74afc7831d39ae0
+size 161533192

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af5891a15588546db1ac7f2baf8fa94835a51a85c032c39793a55bb048b47446
-size 11422523
+oid sha256:ea43b288542655d72d632195ab9b58ca2cd9532c292bf6667827ce899ad196bc
+size 11422082

tokenizer_config.json

@@ -5,7 +5,9 @@
   "clean_up_tokenization_spaces": false,
   "eos_token": "<|im_end|>",
   "errors": "replace",
+  "extra_special_tokens": [],
   "is_local": false,
+  "local_files_only": false,
   "model_max_length": 32768,
   "pad_token": "<|PAD_TOKEN|>",
   "padding_side": "right",