Parcourir la source
fix: next suggest question logic problem (#6451)
Co-authored-by: evenyan <yikun.yan@ubtrobot.com>tags/0.6.15
Even
il y a 1 an
2 fichiers modifiés avec 2 ajouts et 1 suppressions
+ 1
- 0
api/core/llm_generator/prompts.py
Voir le fichier
| SUGGESTED_QUESTIONS_AFTER_ANSWER_INSTRUCTION_PROMPT = ( | SUGGESTED_QUESTIONS_AFTER_ANSWER_INSTRUCTION_PROMPT = ( | ||||
| "Please help me predict the three most likely questions that human would ask, " | "Please help me predict the three most likely questions that human would ask, " | ||||
| "and keeping each question under 20 characters.\n" | "and keeping each question under 20 characters.\n" | ||||
| "MAKE SURE your output is the SAME language as the Assistant's latest response(if the main response is written in Chinese, then the language of your output must be using Chinese.)!\n" | |||||
| "The output must be an array in JSON format following the specified schema:\n" | "The output must be an array in JSON format following the specified schema:\n" | ||||
| "[\"question1\",\"question2\",\"question3\"]\n" | "[\"question1\",\"question2\",\"question3\"]\n" | ||||
| ) | ) |
+ 1
- 1
api/core/memory/token_buffer_memory.py
Voir le fichier
| if curr_message_tokens > max_token_limit: | if curr_message_tokens > max_token_limit: | ||||
| pruned_memory = [] | pruned_memory = [] | ||||
| while curr_message_tokens > max_token_limit and prompt_messages: | |||||
| while curr_message_tokens > max_token_limit and len(prompt_messages)>1: | |||||
| pruned_memory.append(prompt_messages.pop(0)) | pruned_memory.append(prompt_messages.pop(0)) | ||||
| curr_message_tokens = self.model_instance.get_llm_num_tokens( | curr_message_tokens = self.model_instance.get_llm_num_tokens( | ||||
| prompt_messages | prompt_messages |
Chargement…