Llama-3-8B-16K / README.md
winglian's picture
add axolotl tag
1565323 verified
|
raw
history blame
294 Bytes
metadata
datasets:
  - Yukang/LongAlpaca-16k-length
tags:
  - axolotl

This is an extended (16K) context version of LLaMA 3 8B (base, not instruct). Trained for five hours on 8x A6000 GPUs, using the Yukang/LongAlpaca-16k-length dataset.

rope_theta was set to 1000000.0. Trained with Axolotl.