Lines Matching full:for

3  * Default settings for the dokullm plugin
5  * This file defines the default configuration values for the LLM integration plugin.
10  * The API endpoint URL for the LLM service
23  * For security, this should be left empty in the default config and set by the user.
30  * The model identifier to use for text processing
32  * Specifies which LLM model to use for processing requests.
42  * Maximum time to wait for a response from the LLM API before timing out.
43  * Set to 30 seconds by default, which should be sufficient for most requests.
50  * The profile for prompts
52  * Specifies which profile to use for the prompts.
60  * The temperature setting for the LLM
64  * and creative. Default is 0.3 for consistent, high-quality responses.
71  * The top-p (nucleus sampling) setting for the LLM
74  * the output more focused, while higher values (0.5-1.0) allow for more diverse outputs.
75  * Default is 0.8 for a good balance between creativity and coherence.
82  * The top-k setting for the LLM
84  * Limits the number of highest probability tokens considered for each step.
86  * allow for more diverse outputs. Default is 20 for balanced diversity.
93  * The min-p setting for the LLM
95  * Sets a minimum probability threshold for token selection. Tokens with probabilities
148  * Controls whether ChromaDB integration is enabled for document storage and retrieval.
159  * This is used for document storage and retrieval.
178  * The tenant name for ChromaDB organization.
198  * The default collection name for document storage.
208  * The default institution identifier for templates.
219  * Used for generating embeddings for document search.
238  * The model name used for generating text embeddings.
239  * Embeddings are used for semantic search in ChromaDB.