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/oleg-dats/nmt/runs/r2yxjjmy) 
+[<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/oleg-dats/nmt/runs/bs08kkal) 
 
 
 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

@@ -27,13 +27,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "v_proj",
+    "k_proj",
+    "up_proj",
+    "q_proj",
     "gate_proj",
     "o_proj",
-    "down_proj",
-    "q_proj",
-    "up_proj",
-    "k_proj",
-    "v_proj"
+    "down_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 sha256:54ab4a548d241baa01e4b8b83c88ee38524204ccef1fd7a4ab46bdbbb92a0899
+oid sha256:9c735cdd397bf0d33eebbec244280e83c6692be4d7be2148802247de827b901c
 size 1260191096

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:297989e10edd5db84e8c66de54ad4034348eb27c2826ea6e4581f0c42fb83fc3
+oid sha256:0801846a0f3b388085244bd46bf6d23781c6793f9497487004cb2f40f167f0ca
 size 6392