Instructions to use sparklabuofu/MiniWebAgent with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use sparklabuofu/MiniWebAgent with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-4B") model = PeftModel.from_pretrained(base_model, "sparklabuofu/MiniWebAgent") - Notebooks
- Google Colab
- Kaggle
Download ablations/task_centric/training_meta.json from sparklabuofu/MiniWebAgent: direct link, hf CLI and curl.
- Browser
- Download file 642 Bytes
-
https://huggingface.co/sparklabuofu/MiniWebAgent/resolve/main/ablations/task_centric/training_meta.json
- Command line
-
hf download hf://sparklabuofu/MiniWebAgent/ablations/task_centric/training_meta.json
-
curl -L -o training_meta.json https://huggingface.co/sparklabuofu/MiniWebAgent/resolve/main/ablations/task_centric/training_meta.json
642 Bytes
| { | |
| "base_model": "Qwen/Qwen3.5-4B", | |
| "families": [ | |
| "Discrete selection", | |
| "Drag & gesture", | |
| "Form transaction", | |
| "Navigation", | |
| "Out-of-page I/O", | |
| "Reasoning base", | |
| "Text entry" | |
| ], | |
| "training_rows": 26019, | |
| "rows_trained": 26016, | |
| "validation_ce": 0.06781610395206371, | |
| "minutes": 334.7, | |
| "args": { | |
| "adapter": "pooled", | |
| "epochs": 1.0, | |
| "lr": 0.0001, | |
| "rank": 64, | |
| "alpha": 128, | |
| "grad_accum": 8, | |
| "beta": 0.0, | |
| "anchor_ratio": 0.5, | |
| "limit": 0, | |
| "max_steps": 0, | |
| "seed": 0, | |
| "ce_on": "all", | |
| "exclude_variants": "", | |
| "frac": 1.0, | |
| "episode_ids": null, | |
| "init_adapter": null, | |
| "drop_done": false | |
| }, | |
| "gpus": 4 | |
| } |