4dreasoner_v1_cot

This model is a fine-tuned version of Qwen/Qwen3-VL-8B-Instruct on the 4dreasoner_v1_cot_train dataset. It achieves the following results on the evaluation set:

• Loss: 0.2968

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0001
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• total_eval_batch_size: 4
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.3491	0.5215	100	0.3426
0.277	1.0417	200	0.3158
0.246	1.5632	300	0.2992
0.1794	2.0834	400	0.3007
0.1734	2.6050	500	0.2975

Framework versions

• PEFT 0.17.1
• Transformers 4.57.1
• Pytorch 2.11.0+cu130
• Datasets 4.0.0
• Tokenizers 0.22.2