Training in progress, step 1

README.md

@@ -4,8 +4,8 @@ library_name: transformers
 model_name: qwen2-7b-instruct-trl-sft
 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/ruoxue2-stony-brook-university/qwen2vl-sft-mydataset/runs/qigt436p) 
+[<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/ruoxue2-stony-brook-university/qwen2vl-sft-mydataset/runs/1kjq3ipz) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -25,8 +25,8 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "q_proj"
+    "q_proj",
+    "v_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:25ed36143cf7b0ad607488e17722b2e9bae03702a5ccffc367d029a98e6d97da
+oid sha256:b487aa73b73e9133c405eff5c85227b2da3c79b4a3f1f55e98ea993a358d24ca
 size 10110864

eval_dump/ground_truth.jsonl

@@ -0,0 +1 @@
+{"id": "000000497599", "gt": "In this image there are some persons one person is watching the laptop wearing the blue t-shirt and black pant and behind the person the two persons are sitting on the chair and behind the persons there is a big window."}

eval_dump/predictions.jsonl

@@ -0,0 +1 @@
+{"id": "000000497599", "pred": "[GENERATION_ERROR: expected mat1 and mat2 to have the same dtype, but got: float != c10::BFloat16]"}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:04f2388dc9a3d71b93a3f264a16bf12e417e42df37788954d9725f077375a9b5
+oid sha256:83462eb3d7662f8fdb13d576da0f2ff62363d37d3d5e6e17942ada8c07e7b056
 size 5816