Instructions to use MaterialsandtheEnvironment/matenv-smollm2-360m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers.js
How to use MaterialsandtheEnvironment/matenv-smollm2-360m with Transformers.js:
// npm i @huggingface/transformers import { pipeline } from '@huggingface/transformers'; // Allocate pipeline const pipe = await pipeline('text-generation', 'MaterialsandtheEnvironment/matenv-smollm2-360m');
MatEnv SmolLM2-360M
SmolLM2-360M-Instruct fine-tuned with LoRA on questions generated from the data and rules of Materials & the Environment (Chapters 1โ4). It answers directly, or emits a JSON call to the book's exact calculators. Chapter 5 runs it in the browser.
Held-out accuracy (240 questions with unseen phrasings): base {'tool': 0.0, 'direct': 0.158}, fine-tuned {'tool': 0.796, 'direct': 0.721}.
- Downloads last month
- 222
Model tree for MaterialsandtheEnvironment/matenv-smollm2-360m
Base model
HuggingFaceTB/SmolLM2-360M Quantized
HuggingFaceTB/SmolLM2-360M-Instruct