Upload Qwen2VLForConditionalGeneration

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/formless-runs/qwen2-vl-7b-instruct-lora-dpo-merged",
+  "_name_or_path": "/home/andrewhinh/Desktop/Projects/formless/training/artifacts/runs/qwen2-vl-7b-instruct-lora-dpo-merged",
   "architectures": [
     "Qwen2VLForConditionalGeneration"
   ],
@@ -30,8 +30,8 @@
   "rope_theta": 1000000.0,
   "sliding_window": 32768,
   "tie_word_embeddings": false,
-  "torch_dtype": "float32",
-  "transformers_version": "4.46.1",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.45.2",
   "use_cache": true,
   "use_sliding_window": false,
   "video_token_id": 151656,

generation_config.json

@@ -1,4 +1,5 @@
 {
+  "attn_implementation": "flash_attention_2",
   "bos_token_id": 151643,
   "do_sample": true,
   "eos_token_id": [
@@ -9,5 +10,5 @@
   "temperature": 0.01,
   "top_k": 1,
   "top_p": 0.001,
-  "transformers_version": "4.46.1"
+  "transformers_version": "4.45.2"
 }

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0191b87454e636c34c84db80eb1058eb2fafc1ec650878a5442b6f730e9edd6b
+size 4966659944

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18dc376171140ad0dc38bad55470f53f879d3e2f583cd653c0a82316d749ce7f
+size 4991495816

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ebe52972b4ea88a2977d90b2495aab2115e3fdc289b0f7843b290781ca80b88
+size 4932751040

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a021f5f81238a1171b3964aa25e6806b90e0f8862ba693b4e129b97cce2e6a6
+size 1691924384