Transformers Text-Generation Pipeline Q&A: Adapter vs. Base Model

A tool for experimenting with PEFT (Parameter-Efficient Fine-tuning) and LoRA (Low-Rank Adaptation) adapter performance.

Select an adapter or enter a new adapter

joshause/llama-3.2-1b-dewey-how-we-think-adapter joshause/gemma-3-1b-pt-spinoza-treatise-emendation-intellect Other

Base model (auto-filled)

Select a question or enter a new question

SAMPLE QUESTION: Does the algorithmic undecidability in the Halting Problem reflect metaphysical truths about reality and real limits of physical computation as suggested by Turing, or does it instead reveal nothing more than a boundary of the rules of language and the way we use language as suggested by Wittgenstein's rule-following paradox? TRAINING QUESTION: Why does Spinoza consider doubt a signal of an inadequate idea? TRAINING QUESTION: How does Dewey describe the interplay between analysis and synthesis during judgment? NEW QUESTION: Other

Max new tokens

max_new_tokens in text-generation pipeline limits the length of the generated answer

1 512

Temperature

temperature in text-generation pipeline parameter controls the randomness of the generated text

0.1 0.9

Adapter output

Base model output