Model save

README.md

@@ -1,13 +1,9 @@
 ---
 base_model: Qwen/Qwen2.5-VL-3B-Instruct
-datasets: smolagents/aguvis-stage-2
 library_name: transformers
 model_name: Qwen2.5-VL-3B-Instruct-Agentic
 tags:
 - generated_from_trainer
-- open-r1
-- vision-language
-- vlm
 - trl
 - sft
 licence: license
@@ -15,7 +11,7 @@ licence: license
 
 # Model Card for Qwen2.5-VL-3B-Instruct-Agentic
 
-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 [smolagents/aguvis-stage-2](https://huggingface.co/datasets/smolagents/aguvis-stage-2) 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
@@ -31,7 +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/huggingface/huggingface/runs/mkymtipi) 
+[<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/huggingface/huggingface/runs/a5t7btov) 
 
 
 This model was trained with SFT.
@@ -41,8 +37,8 @@ This model was trained with SFT.
 - TRL: 0.18.0
 - Transformers: 4.52.3
 - Pytorch: 2.6.0
-- Datasets: 3.6.0
-- Tokenizers: 0.21.1
+- Datasets: 4.0.0
+- Tokenizers: 0.21.2
 
 ## Citations
 

all_results.json

@@ -1,8 +1,8 @@
 {
-    "total_flos": 309043048284160.0,
-    "train_loss": 0.11322491052526659,
-    "train_runtime": 3926.7142,
-    "train_samples": 20495,
-    "train_samples_per_second": 31.316,
-    "train_steps_per_second": 0.49
+    "total_flos": 1.1033250327691264e+16,
+    "train_loss": 0.6782059136778116,
+    "train_runtime": 78870.76,
+    "train_samples": 327501,
+    "train_samples_per_second": 4.152,
+    "train_steps_per_second": 0.008
 }

config.json

@@ -70,7 +70,7 @@
   },
   "torch_dtype": "bfloat16",
   "transformers_version": "4.52.3",
-  "use_cache": true,
+  "use_cache": false,
   "use_sliding_window": false,
   "video_token_id": 151656,
   "vision_config": {

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:736013d0a86054665b4246806a0307536ccbb0b71d1ff03895e2fab630a64c18
+oid sha256:50c6779f3e08adbe8eafe5bf9c1f48b70f50c4d8a78e1fecbcfe19c4a3c1e140
 size 4997756664

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c17f0453a22b53ee5311492941a8abc7cd6ec9e6a0cd2358ff3b1e237b9afa60
+oid sha256:1b59679cf1cb5e5267415e940d8158c1d1e1a5a9192116ce281aed6294eba115
 size 2511590128

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba0c439f7be467bf47d12a7e6f9adc6116201056fc60c67f431c679b7c16afc8
-size 11422064
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

tokenizer_config.json

@@ -199,9 +199,10 @@
   "eos_token": "<|im_end|>",
   "errors": "replace",
   "extra_special_tokens": {},
+  "max_pixels": 1003520,
+  "min_pixels": 200704,
   "model_max_length": 131072,
   "pad_token": "<|endoftext|>",
-  "processor_class": "Qwen2_5_VLProcessor",
   "split_special_tokens": false,
   "tokenizer_class": "Qwen2Tokenizer",
   "unk_token": null

train_results.json

@@ -1,8 +1,8 @@
 {
-    "total_flos": 309043048284160.0,
-    "train_loss": 0.11322491052526659,
-    "train_runtime": 3926.7142,
-    "train_samples": 20495,
-    "train_samples_per_second": 31.316,
-    "train_steps_per_second": 0.49
+    "total_flos": 1.1033250327691264e+16,
+    "train_loss": 0.6782059136778116,
+    "train_runtime": 78870.76,
+    "train_samples": 327501,
+    "train_samples_per_second": 4.152,
+    "train_steps_per_second": 0.008
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a173f08d1eae91c16a9c6221f280e019a42ffd0ba8c5ded0e47d5670fa39f3b
-size 7352
+oid sha256:4c0ce0e66c77eae20e729b7f22ffbfec3eec08941b70655111f37221a08e545b
+size 7416