Meta-Llama-3-12B-Instruct is a depth upscaling merge of llama3-8b from M. Labonne
274 Pulls Updated 6 months ago
Updated 6 months ago
6 months ago
ab100bc8cdd0 · 3.8GB
model
archllama
·
parameters11.5B
·
quantizationIQ4_XS
3.8GB
params
{
"num_keep": 24,
"stop": [
"<|start_header_id|>",
"<|end_header_id|>",
110B
template
{{ if .System }}<|start_header_id|>system<|end_header_id|>
{{ .System }}<|eot_id|>{{ end }}{{ if
260B
Readme
Meta-Llama-3-12B-Instruct is a depth upscaling merge of llama3-8b from M. Labonne
Quantized with imatrix using groups_merged.txt