Training in progress, step 100

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/tphage99/huggingface/runs/yc7yg1zg) 
+[<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/tphage99/huggingface/runs/m49behse) 
 
 
 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

@@ -24,13 +24,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "k_proj",
     "gate_proj",
+    "down_proj",
     "o_proj",
     "q_proj",
-    "up_proj",
-    "down_proj"
+    "k_proj",
+    "v_proj",
+    "up_proj"
   ],
   "task_type": "CAUSAL_LM",
   "trainable_token_indices": null,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b584d89a766f1934090792d94fb823fbfa774948a74f0ac7180393745646183e
+oid sha256:b7b08ea947219348e7d7621c75af27269d0ae443cb9c51f6fd8859f5880ca9fe
 size 97307544

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be1220ce9c0ebcc7ed8c513a78357b955cba140f084520763156784779ed6316
+oid sha256:98a227c5e63e5ea2bbc9d0ecc3253dd099f6bedbbff3bf6a601b5dd5fabd20d6
 size 6456