Initial commit

config.json

@@ -1,5 +1,4 @@
 {
-  "_name_or_path": "../checkpoints/openvlthinker_grpo_iter3_hard/global_step_30/actor/huggingface",
   "architectures": [
     "Qwen2_5_VLForConditionalGeneration"
   ],
@@ -31,17 +30,33 @@
   "sliding_window": 32768,
   "tie_word_embeddings": false,
   "torch_dtype": "float32",
-  "transformers_version": "4.49.0",
+  "transformers_version": "4.51.0",
   "use_cache": false,
   "use_sliding_window": false,
   "video_token_id": 151656,
   "vision_config": {
+    "depth": 32,
+    "fullatt_block_indexes": [
+      7,
+      15,
+      23,
+      31
+    ],
+    "hidden_act": "silu",
     "hidden_size": 1280,
+    "in_channels": 3,
     "in_chans": 3,
+    "intermediate_size": 3420,
     "model_type": "qwen2_5_vl",
+    "num_heads": 16,
+    "out_hidden_size": 3584,
+    "patch_size": 14,
+    "spatial_merge_size": 2,
     "spatial_patch_size": 14,
+    "temporal_patch_size": 2,
     "tokens_per_second": 2,
-    "torch_dtype": "float32"
+    "torch_dtype": "float32",
+    "window_size": 112
   },
   "vision_end_token_id": 151653,
   "vision_start_token_id": 151652,

generation_config.json

@@ -2,6 +2,6 @@
   "_from_model_config": true,
   "eos_token_id": 151645,
   "pad_token_id": 151643,
-  "transformers_version": "4.49.0",
+  "transformers_version": "4.51.0",
   "use_cache": false
 }

model-00001-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e049b9e8e895ab98948e5dda7c7677e7030bcf3fa6250fa306802d570ceccfa7
-size 4940083336
+oid sha256:43a1cc3c3c5582d0ec4f21ecc0ec1da402ea2fcdddc83c89ab3c4c4793bed11a
+size 4927980624

model-00002-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3f6856a1ab07a9939ce9a60380fc76986d7bc6f9da6943e731d1f4b116744497
-size 4939017568
+oid sha256:f18201c56785c99564e941cb0653d4089c6dd20d3130d8e1f7039f9b65b648c2
+size 4977500472

model-00003-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1b64f0da32053c2e7ba8b4b31425b154998a058c0d3d16beb651203c22ee5911
-size 4350476240
+oid sha256:c1667b3d1e2f67f07ce72a782e6970960dfe8584d74f1720a730e06eda8c1059
+size 4906314960

model-00004-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bcd688eec389c6faf95b80a57300b64842820e6b8e16cb044014f5abee5f6ed7
-size 2354837408
+oid sha256:b5c8da21324e49ea35e7649ba96126f76233cccd8f7b2ee6b0a1230d8b3d599d
+size 1772618480

preprocessor_config.json

@@ -1,6 +1,8 @@
 {
   "crop_size": null,
+  "data_format": "channels_first",
   "default_to_square": true,
+  "device": null,
   "do_center_crop": null,
   "do_convert_rgb": true,
   "do_normalize": true,
@@ -17,6 +19,7 @@
     0.26130258,
     0.27577711
   ],
+  "input_data_format": null,
   "max_pixels": 12845056,
   "merge_size": 2,
   "min_pixels": 3136,
@@ -24,6 +27,7 @@
   "processor_class": "Qwen2_5_VLProcessor",
   "resample": 3,
   "rescale_factor": 0.00392156862745098,
+  "return_tensors": null,
   "size": {
     "longest_edge": 12845056,
     "shortest_edge": 3136