Training in progress, step 20

README.md

@@ -4,8 +4,8 @@ library_name: transformers
 model_name: medgemma-4b-it-sft-lora-be
 tags:
 - generated_from_trainer
-- trl
 - sft
+- 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/wangyichen25-mayo-clinic/medgemma_finetune/runs/7emypqp3) 
+[<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/wangyichen25-mayo-clinic/medgemma_finetune/runs/tnqd3ddd) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -28,16 +28,16 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "k_proj",
-    "gate_proj",
     "fc1",
+    "o_proj",
+    "fc2",
+    "out_proj",
     "q_proj",
-    "v_proj",
+    "gate_proj",
     "up_proj",
     "down_proj",
-    "out_proj",
-    "fc2"
+    "v_proj",
+    "k_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 sha256:36a1e370f0218a3d977396c210a401c534230ad4c063e4703b270fdb365b1eff
+oid sha256:cdda4478dbd5bd2e9d6fd020c8f89c04cf43e9c7cb839ca8e27ec76a4814e478
 size 2839126480

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2661e9d16a322a8b69865bc4b33ea73e1b304a23fa93839c76bab58d65645833
+oid sha256:652c8ce42c66ba23561de837fae1da08a712795400ecc170511dcf7fa6b44940
 size 5816