General use chat model based on Llama and Llama 2 with 2K to 16K context sizes。
7b 13b 33b
176.2K Pulls Updated 16 months ago
Updated 16 months ago
16 months ago
c85a3d3b3b78 · 9.2GB
model
archllama
·
parameters13B
·
quantizationQ5_K_M
9.2GB
params
{ "stop": [ "USER:", "ASSISTANT:" ] }
31B
template
{{ .System }} USER: {{ .Prompt }} ASSISTANT:
45B
system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful,
155B
Readme
Vicuna 是一個聊天助理模型。它包含 3 種不同變體,各有 3 種不同尺寸。v1.3 是通過微調 Llama 訓練的,上下文大小為 2048 個 tokens。v1.5 是通過微調 Llama 2 訓練的,上下文大小為 2048 個 tokens。v1.5-16k 是通過微調 Llama 2 訓練的,上下文大小為 16k 個 tokens。所有三種變體都使用從 ShareGPT 收集的對話進行訓練。
Example prompts
What is the meaning of life? Explain it in 5 paragraphs.