Ousso1117/GRPO-meta-Llama-3.2-3B-meta-Llama-3.2-3B-mrd3-sum

README.md

@@ -27,7 +27,7 @@ print(output["generated_text"])
 
 ## Training procedure
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/mnlp_osy/huggingface/runs/9xtc3ahe) 
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/mnlp_osy/huggingface/runs/m41g55y6) 
 
 
 This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).

adapter_config.json

@@ -23,14 +23,14 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "up_proj",
-    "o_proj",
-    "down_proj",
-    "k_proj",
     "ff_proj",
-    "mlp_proj",
+    "k_proj",
     "v_proj",
     "q_proj",
+    "down_proj",
+    "mlp_proj",
+    "o_proj",
+    "up_proj",
     "gate_proj"
   ],
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9077494471c4d78d8778170d52b913a69797f0ca7134d414cc9a2b5c7f81f001
+oid sha256:8b0af35769ab3a2c845f04eece2054cad030048f555654b5d6ac037dd5898090
 size 97307936

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:309bc3cc194c927b08f50ff10bd4886e877df23d9e4dfff1627967de9d05f910
+oid sha256:fff6920dd6bfc40916a46be0e857f28870fa0232fa3f33685760dad2c6754ba8
 size 6648