Training in progress, step 10

README.md

@@ -1,6 +1,5 @@
 ---
 base_model: Qwen/Qwen2.5-VL-3B-Instruct
-datasets: lmms-lab/multimodal-open-r1-8k-verified
 library_name: transformers
 model_name: Qwen2.5-VL-3B-Instruct-Thinking
 tags:
@@ -12,7 +11,7 @@ licence: license
 
 # Model Card for Qwen2.5-VL-3B-Instruct-Thinking
 
-This model is a fine-tuned version of [Qwen/Qwen2.5-VL-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct) on the [lmms-lab/multimodal-open-r1-8k-verified](https://huggingface.co/datasets/lmms-lab/multimodal-open-r1-8k-verified) dataset.
+This model is a fine-tuned version of [Qwen/Qwen2.5-VL-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ecb1b194f36d9e7c1f2faad94c41d7d44ed589ff49072bc9ded628100514c0bd
+oid sha256:16ad78ab6fa8ca02ae2d84afa2379bdc31da86eeab1529ca3f0f23c7d64a4120
 size 7393888

preprocessor_config.json

@@ -1,4 +1,10 @@
 {
+  "crop_size": null,
+  "data_format": "channels_first",
+  "default_to_square": true,
+  "device": null,
+  "disable_grouping": null,
+  "do_center_crop": null,
   "do_convert_rgb": true,
   "do_normalize": true,
   "do_rescale": true,
@@ -8,12 +14,13 @@
     0.4578275,
     0.40821073
   ],
-  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_processor_type": "Qwen2VLImageProcessorFast",
   "image_std": [
     0.26862954,
     0.26130258,
     0.27577711
   ],
+  "input_data_format": null,
   "max_pixels": 12845056,
   "merge_size": 2,
   "min_pixels": 3136,
@@ -21,6 +28,7 @@
   "processor_class": "Qwen2_5_VLProcessor",
   "resample": 3,
   "rescale_factor": 0.00392156862745098,
+  "return_tensors": null,
   "size": {
     "longest_edge": 12845056,
     "shortest_edge": 3136

runs/Jul28_13-21-17_f76ea2c310ec/events.out.tfevents.1753708949.f76ea2c310ec.36632.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b99b1cc8d7053ed056eab6a48c538db5e4c33edf5eb40fa7c1ff940cee741c4f
+size 10601

tokenizer_config.json

@@ -205,5 +205,6 @@
   "processor_class": "Qwen2_5_VLProcessor",
   "split_special_tokens": false,
   "tokenizer_class": "Qwen2Tokenizer",
-  "unk_token": null
+  "unk_token": null,
+  "use_fast": true
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0fc252ac692415b04381a46218fbb5ea43ecd5259a89189903fa9f64f9f23e9d
-size 6520
+oid sha256:43028820679d4f6ee87f1f232ec6907e43bcdaa245be62b39aec0f5de9c20c4e
+size 6584