Models
GitHub
Discord
Turbo
Sign in
Download
Models
Download
GitHub
Discord
Sign in
aisingapore
/
Llama-SEA-LION-v3.5-70B-R
:q5_k_m
40
Downloads
Updated
4 months ago
Llama-SEA-LION-v3.5-70B-R is a multilingual model which has been pretrained and instruct-tuned for the Southeast Asia region. Developed by AI Singapore and funded by National Research Foundation, Singapore.
Llama-SEA-LION-v3.5-70B-R is a multilingual model which has been pretrained and instruct-tuned for the Southeast Asia region. Developed by AI Singapore and funded by National Research Foundation, Singapore.
Cancel
Llama-SEA-LION-v3.5-70B-R:q5_k_m
...
/
template
397e657fdb3d · 880B
{{- /* Begin with the beginning of sequence token if present */}}
<|begin▁of▁sentence|>
{{- /* Since thinking_mode is always "on", we include this line */}}
detailed thinking on
{{- /* Output the system prompt if present */}}
{{- if .System }}
{{ .System }}
{{- else }}
{{- /* Collect and output system messages from .Messages if any */}}
{{- range .Messages }}
{{- if eq .Role "system" }}
{{ .Content }}
{{- end }}
{{- end }}
{{- end }}
{{- /* Iterate over the messages */}}
{{- range .Messages }}
{{- if eq .Role "user" }}
<|User|>{{ .Content }}
{{- else if eq .Role "assistant" }}
{{- if .Content }}
{{- /* Output the assistant's content */}}
<|Assistant|><think>\n{{ .Content }}<|end▁of▁sentence|>
{{- end }}
{{- end }}
{{- end }}
{{- /* Always add the generation prompt at the end */}}
<|Assistant|><think>\n