rixty
/
SillyTavern
mirror of https://github.com/SillyTavern/SillyTavern.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
SillyTavern/public/instruct/Llama2.json

15 lines
420 B
JSON
Raw Normal View History

Created Llama2 instruct preset Tested on Llama 2 13B chat model provided by TheBloke. Based on these prompt example: https://huggingface.co/TheBloke/Llama-2-13B-chat-GPTQ#prompt-template-llama-2-chat
2023-07-22 09:24:17 +02:00
{
"name": "Llama 2",
Fixed tags order in Llama2 instruct preset I didn't notice that the prompt wording implies impersonation, which is incorrect for common usage. Changed the order of tags used to generate the character's response to the user, rather than the vice versa.
2023-07-22 17:31:44 +02:00
"system_prompt": "Write {{char}}'s next reply in this fictional roleplay with {{user}}.\n<</SYS>>\n",
Created Llama2 instruct preset Tested on Llama 2 13B chat model provided by TheBloke. Based on these prompt example: https://huggingface.co/TheBloke/Llama-2-13B-chat-GPTQ#prompt-template-llama-2-chat
2023-07-22 09:24:17 +02:00
"input_sequence": "[INST]",
"output_sequence": "[/INST]",
Add Last output sequence setting for instruct mode
2023-08-17 22:33:27 +02:00
"last_output_sequence": "",
Added Include Names and Forced Names to all instruct presets
2023-08-20 16:10:22 +02:00
"system_sequence": "[INST] <<SYS>>\n",
"stop_sequence": "",
Created Llama2 instruct preset Tested on Llama 2 13B chat model provided by TheBloke. Based on these prompt example: https://huggingface.co/TheBloke/Llama-2-13B-chat-GPTQ#prompt-template-llama-2-chat
2023-07-22 09:24:17 +02:00
"separator_sequence": "\n",
Added Include Names and Forced Names to all instruct presets
2023-08-20 16:10:22 +02:00
"wrap": false,
"macro": true,
"names": false,
"names_force_groups": true
Created Llama2 instruct preset Tested on Llama 2 13B chat model provided by TheBloke. Based on these prompt example: https://huggingface.co/TheBloke/Llama-2-13B-chat-GPTQ#prompt-template-llama-2-chat
2023-07-22 09:24:17 +02:00
}