huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-12 12:54:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix: add template test and update docs

Browse Source
This commit is contained in:
drbh 2024-08-09 14:16:35 +00:00
parent 3b25cd3213
commit d94b0fcf52
2 changed files with 20 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
docs/openapi.json
View File

@ -819,6 +819,13 @@
"example": "1.0",
"nullable": true
},
"guideline": {
"type": "string",
"description": "A guideline to be used in the chat_template",
"default": "null",
"example": "null",
"nullable": true
},
"logit_bias": {
"type": "array",
"items": {

13
router/src/infer/chat_template.rs
View File

@ -733,6 +733,19 @@ mod tests {
},
target: "You are a friendly chatbot who always responds in the style of a pirateYou are a friendly chatbot who always responds in the style of a pirate### Instruction: Hello, how are you?### Response: I'm doing great. How can I help you today?### Instruction: I'd like to show off how chat templating works!",
},
ChatTemplateTestItem {
name: "google/shieldgemma-9b",
chat_template: "{{- bos_token }}\n{%- if messages[-1].role == 'user' %}\n {{- \"You are a policy expert trying to help determine whether a user prompt is in violation of the defined safety policies.\\n\\n\" }}\n {{- \"<start_of_turn>\\n\" }}\n {{- \"Human Question: \" + messages[-1].content }}\n {{- \"\\n<end_of_turn>\\n\\n\" }}\n {{- \"Our safety principle is defined in the below:\\n\\n\" }}\n {{- \"* \" + guideline + \"\\n\" }}\n {{- \"\\n===\\n\\n\" }}\n {{- \"Does the human question violate the above principle? Your answer must start with 'Yes' or 'No'. And then walk through step by step to be sure we answer correctly.\\n\\n\" }}\n{%- else %}\n {{- \"You are a policy expert trying to help determine whether an AI response to a prompt is in violation of the defined safety policies.\\n\\n\" }}\n {{- \"<start_of_turn>\\n\" }}\n {{- \"Human Question: \" + messages[-2].content }}\n {{- \"\\n<end_of_turn>\\n\" }}\n {{- \"<start_of_turn>\\n\" }}\n {{- \"Chatbot Response: \" + messages[-1].content }}\n {{- \"\\n<end_of_turn>\\n\\n\" }}\n {{- \"Our safety principle is defined in the below:\\n\\n\" }}\n {{- \"* \" + guideline + \"\\n\" }}\n {{- \"\\n===\\n\\n\" }}\n {{- \"Does the Chatbot Response violate the above principle? Your answer must start with 'Yes' or 'No'. And then walk through step by step to be sure we answer correctly.\\n\\n\" }}\n{%- endif %}\n\n",
input: ChatTemplateInputs {
messages: example_chat_with_system.clone(),
add_generation_prompt: false,
bos_token: Some("<s>"),
eos_token: Some("</s>"),
guideline: Some("Do not use offensive language."),
..Default::default()
},
target: "<s>You are a policy expert trying to help determine whether a user prompt is in violation of the defined safety policies.\n\n<start_of_turn>\nHuman Question: I'd like to show off how chat templating works!\n<end_of_turn>\n\nOur safety principle is defined in the below:\n\n* Do not use offensive language.\n\n===\n\nDoes the human question violate the above principle? Your answer must start with 'Yes' or 'No'. And then walk through step by step to be sure we answer correctly.\n\n",
},
];
#[allow(unused_variables)] // name is unused