ai_toxicity_inference_service_api_endpoint:"URL where the API is running for the toxicity module"
ai_toxicity_inference_service_api_key:"API key for the toxicity API"
ai_toxicity_inference_service_api_model:"Model to use for inference. Multilingual model works with Italian, French, Russian, Portuguese, Spanish and Turkish."
ai_toxicity_flag_automatically:"Automatically flag posts / chat messages that are above the configured thresholds."
ai_toxicity_flag_threshold_toxicity:"Toxicity: a rude, disrespectful, or unreasonable comment that is somewhat likely to make you leave a discussion or give up on sharing your perspective"
ai_toxicity_flag_threshold_severe_toxicity:"Severe Toxicity: a very hateful, aggressive, or disrespectful comment that is very likely to make you leave a discussion or give up on sharing your perspective"
ai_toxicity_groups_bypass:"Users on those groups will not have their posts classified by the toxicity module."
ai_sentiment_enabled:"Enable the sentiment module."
ai_sentiment_inference_service_api_endpoint:"URL where the API is running for the sentiment module"
ai_sentiment_inference_service_api_key:"API key for the sentiment API"
ai_sentiment_models:"Models to use for inference. Sentiment classifies post on the positive/neutral/negative space. Emotion classifies on the anger/disgust/fear/joy/neutral/sadness/surprise space."
ai_openai_gpt35_url:"Custom URL used for GPT 3.5 chat completions. (Azure: MUST support function calling and ideally is a GPT3.5 16K endpoint)"
ai_openai_gpt4_url:"Custom URL used for GPT 4 chat completions. (for Azure support)"
ai_openai_embeddings_url:"Custom URL used for the OpenAI embeddings API. (in the case of Azure it can be: https://COMPANY.openai.azure.com/openai/deployments/DEPLOYMENT/embeddings?api-version=2023-05-15)"
ai_embeddings_model:"Use all-mpnet-base-v2 for local and fast inference in english, text-embedding-ada-002 to use OpenAI API (need API key) and multilingual-e5-large for local multilingual embeddings"
ai_embeddings_pg_connection_string:"PostgreSQL connection string for the embeddings module. Needs pgvector extension enabled and a series of tables created. See docs for more info."