In order of interest:
- assistant_plain: AI assistant with single-line replies.
- Minimial prompt that should work for any model.
- roshambo_kira: Play against a Kira in roshambo.
- Demonstrates why LLMs are hard to get right.
- confidant_alpaca: A camelid that occasionally spits.
- Demonstrates a method of prompting instruction-tuned models to fill in character dialogue.
- Instruction-following AI assistants.
- For all of these examples, the assistant must end its messages with a special token like EOS.
- assistant_alpaca: Alpaca prompt format.
- assistant_chatml: ChatML prompt format that typically requires special
<|im_start|>and<|im_end|>tokens but is configured with fallbacks. - assistant_gemma: Gemma prompt format that requires special
<start_of_turn>and<end_of_turn>tokens. - assistant_llama: Llama 3 prompt format that requires special
<|start_header_id|>,<|end_header_id|>, and<|eot_id|>tokens. - assistant_mistral: Mistral propmt format that requires special
[INST]and[/INST]tokens. - assistant_vicuna: Vicuna prompt format.
- assistant_coprocess: A simple assistant that can be controlled as a coprocess.
- Demonstrates the
/putsand/getscommands.
- Demonstrates the