|
|
--- |
|
|
license: llama2 |
|
|
language: |
|
|
- en |
|
|
pipeline_tag: conversational |
|
|
tags: |
|
|
- merge |
|
|
--- |
|
|
|
|
|
This quant was made for [infermatic.ai](https://infermatic.ai/) |
|
|
|
|
|
Dynamic FP8 quant of [goliath-120b-Instruct-FP8-Dynamic](https://huggingface.co/alpindale/goliath-120b) made with AutoFP8. |
|
|
|
|
|
|
|
|
# Goliath 120B |
|
|
|
|
|
An auto-regressive causal LM created by combining 2x finetuned [Llama-2 70B](https://huggingface.co/meta-llama/llama-2-70b-hf) into one. |
|
|
|
|
|
# Prompting Format |
|
|
|
|
|
Both Vicuna and Alpaca will work, but due the initial and final layers belonging primarily to Xwin, I expect Vicuna to work the best. |
