Training in progress, step 5

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/g-puca1-deloitte/llmv3/runs/74wqm5co) 
+[<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/g-puca1-deloitte/llmv3/runs/srxf8mv6) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -25,10 +25,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
-    "q_proj",
     "v_proj",
-    "o_proj"
+    "k_proj",
+    "o_proj",
+    "q_proj"
   ],
   "target_parameters": [
     "0.mlp.experts.gate_up_proj",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:78be3e6ca5358911aa5ab362d8ec99d9539c2144087af3e901b622d8dc92c793
+oid sha256:eb8da8652a20fad6bc397d83b3e8c18df5bd7610da101adcd877516f60747921
 size 200875760

modelopt_state_train.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:abf87c2c23f963b7b41dad262ae79da7e98d2afd11ef201b690dbce5dcfdcf0f
+oid sha256:e681e53732d85f035c2d0312b9f9faa3a6c78705c7c7f6181a6f63a08c891fdf
 size 994683

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:caa41525d7bc490cee9ba84c5abc4563bdf5d0e8fa696b5e713dc8159d306c55
+oid sha256:3161d867e1ae44b588b91075f3118b054bee7d47a03c6ee76c4479237a4b2a55
 size 6289