LLM Speculative Decoding - a Doctor-Shotgun Collection

Doctor-Shotgun 's Collections

Stable Diffusion Things

LLMs

LLM Speculative Decoding

LLM Speculative Decoding

updated Aug 31

Tiny language models meant to serve as draft models for speculative decoding.

Doctor-Shotgun/TinyLlama-1.1B-32k

Text Generation • Updated Feb 2 • 123 • 28
Doctor-Shotgun/TinyLlama-1.1B-32k-Instruct

Text Generation • Updated Jun 7 • 4.93k • 12
Doctor-Shotgun/smol_llama-220M-GQA-32k-theta

Text Generation • Updated Jan 10 • 18 • 1
Doctor-Shotgun/smol_llama-220M-GQA-32k-theta-sft

Text Generation • Updated Jan 10 • 14 • 2
Doctor-Shotgun/smol_llama-220M-GQA-32k-theta-sft-limarp

Text Generation • Updated Jan 10 • 12
Doctor-Shotgun/smol_llama-220M-GQA-32k-linear

Text Generation • Updated Jan 10 • 16