Training in progress, step 380

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/rl-research-team/curvebench/runs/2l0unhzs) 
+[<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/rl-research-team/curvebench/runs/rk873hvr) 
 
 
 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

@@ -25,16 +25,16 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "fc2",
-    "up_proj",
-    "fc1",
-    "o_proj",
-    "k_proj",
-    "gate_proj",
     "out_proj",
     "v_proj",
+    "k_proj",
     "down_proj",
-    "q_proj"
+    "gate_proj",
+    "fc2",
+    "up_proj",
+    "fc1",
+    "q_proj",
+    "o_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:b741c6a7dbbdc1219401db11960b68719511625b1244c6a1b734d136debeec68
+oid sha256:0e0aac2cebf10dda055c906219fdf2fdbf76c300479644f9e32c90bb277a6a7d
 size 74318104

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7b344ba41e0ad4c00487059aecbc3ce36ba9daf367f8f39a07cd7b5ee0ece2df
+oid sha256:9b104f4880b5e5596f32303633dee230104c79c5efffe268b737eb82092f0cff
 size 7569