Training in progress, step 500

README.md

@@ -4,8 +4,8 @@ library_name: transformers
 model_name: Qwen2.5-VL-3B-Instruct-Thinking
 tags:
 - generated_from_trainer
-- trl
 - grpo
+- trl
 licence: license
 ---
 

adapter_config.json

@@ -25,13 +25,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "v_proj",
     "k_proj",
-    "gate_proj",
     "up_proj",
-    "down_proj",
-    "q_proj"
+    "o_proj",
+    "gate_proj",
+    "v_proj",
+    "q_proj",
+    "down_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:64f0dd95f25dacc6dfcc1de1a33c38aeae82bfac6b62fdec821fa0e6b1b93993
+oid sha256:a201eb79b883696f3308cfe084c98f161ba392b9f7a64c708ef71e68ce807a0a
 size 74405904

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9c672c52b27e6be4e079b4efe127c5f583aebaa9ce72ff9e263230f1e7f72d8e
+oid sha256:0e736039d09a02f279c14a2b6fe16882e7396e8246b1637223973da8a0041ca6
 size 7185