Training in progress, step 10

README.md

@@ -4,8 +4,8 @@ library_name: transformers
 model_name: glv4-1v-all-linear-500sample-random
 tags:
 - generated_from_trainer
-- trl
 - grpo
+- trl
 licence: license
 ---
 
@@ -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/on-planets/glm4-1v-8b/runs/oy234co0) 
+[<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/on-planets/glm4-1v-8b/runs/x9wg06f1) 
 
 
 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

@@ -25,17 +25,17 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "up_proj",
     "gate_up_proj",
-    "k_proj",
+    "merger.proj",
     "q_proj",
-    "gate_proj",
     "qkv",
-    "v_proj",
     "attn.proj",
-    "merger.proj",
-    "up_proj",
-    "down_proj",
-    "o_proj"
+    "gate_proj",
+    "k_proj",
+    "o_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:b5d73296ea86b17a38f65541f5136473df01899274eae985c8cd8271670428af
+oid sha256:66ca24b714e78d6918189fcc45b8b94db809e951e01277ba9867521f13162441
 size 117320232

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:185a61227ee260ef096ee964c230d7fca7146b2982090201346ca40037bb2c0a
+oid sha256:ad6cbdceb997e880e095b8f23c8ba5d1da3757b205edce00c92a8240a2a73018
 size 6993