duoan/siq-vl_siglip2-large-patch16-512_qwen2.5-1.5b-instruct_stage2

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

siq-vlsiglip2-large-patch16-512qwen2.5-1.5b-instruct_stage2

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• —Loss: 1.5344
• —Num Input Tokens Seen: 107807747

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: 2e-05
• —trainbatchsize: 4
• —evalbatchsize: 4
• —seed: 42
• —distributed_type: multi-GPU
• —num_devices: 8
• —gradientaccumulationsteps: 4
• —totaltrainbatch_size: 128
• —totalevalbatch_size: 32
• —optimizer: Use adamwtorchfused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• —lrschedulertype: cosine
• —lrschedulerwarmup_ratio: 0.03
• —training_steps: 264503

Training results

Framework versions

• —Transformers 4.57.3
• —Pytorch 2.9.1+cu130
• —Datasets 4.4.1
• —Tokenizers 0.22.1