Text Generation
PEFT
Safetensors
lora
gemma-3
adhd
neurodivergent
task-initiation
build-small-hackathon
backyard-ai
modal
zerogpu
conversational
Instructions to use build-small-hackathon/NeuroBait with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use build-small-hackathon/NeuroBait with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("unsloth/gemma-3-12b-it") model = PeftModel.from_pretrained(base_model, "build-small-hackathon/NeuroBait") - Notebooks
- Google Colab
- Kaggle
| { | |
| "base_model": "unsloth/gemma-3-12b-it", | |
| "train_file": "/data/train.jsonl", | |
| "eval_file": "/data/eval.jsonl", | |
| "out_adapter": "/out/neurobait-lora-12b", | |
| "out_dir": "/out/outputs", | |
| "epochs": 3.0, | |
| "max_seq": 2048, | |
| "seed": 42, | |
| "n_train": 270, | |
| "n_eval": 30, | |
| "train_loss": 1.7501190456689573, | |
| "eval": { | |
| "eval_loss": 1.8844258785247803, | |
| "eval_runtime": 20.1028, | |
| "eval_samples_per_second": 1.492, | |
| "eval_steps_per_second": 1.492, | |
| "epoch": 3.0 | |
| }, | |
| "expected_steps": 102 | |
| } |