Hugging Face's logo

allknowingroger
/
SmolVLM-Base-vqav2

PEFT
TensorBoard
Safetensors
Generated from Trainer
Model card Files Metrics Community
SmolVLM-Base-vqav2 / README.md
allknowingroger's picture
allknowingroger
Update README.md
e4c1124 verified
preview code
|
raw
history blame
1.38 kB
metadata
library_name: peft
license: apache-2.0
base_model: HuggingFaceTB/SmolVLM-Base
tags:
  - generated_from_trainer
model-index:
  - name: SmolVLM-Base-vqav2
    results: []
datasets:
  - osunlp/Multimodal-Mind2Web

SmolVLM-Base-vqav2

This model is a fine-tuned version of HuggingFaceTB/SmolVLM-Base on osunlp/Multimodal-Mind2Web dataset.

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: 4
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 16
  • optimizer: Use paged_adamw_8bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 50
  • num_epochs: 1

Framework versions

  • PEFT 0.13.2
  • Transformers 4.46.2
  • Pytorch 2.5.1+cu121
  • Datasets 3.1.0
  • Tokenizers 0.20.3