0.3.2-understand-kg_entity

[0.4.4] The transient token budget allocated for active reasoning within a single model call; scratch space for chain_

[0.3.5] The schema layer of a knowledge_graph defining allowed kg_entity types (classes), allowed kg_predicate labels,

[0.3.6] A kg_entity type within the kg_ontology; groups instances that share structural or semantic properties

[0.4.1] The finite token sequence an LLM can process in a single forward pass; defines the maximum input + output capa

[0.4.2] Total number of tokens (input + output) an LLM supports; the hard ceiling on how much text can be loaded into

[0.4.3] Information persisted across sessions, tasks, or deployments; requires external storage and retrieval mechanis

[0.2.1] The positional delimiter marking the transition from input context to output generation; tokens before this ma

[0.2.2] The portion of context_length reserved and available for prompt tokens, retrieved context, and system instruct

[0.2.3] The portion of context_length reserved for generated tokens; equals context_length minus input_token_budget mi

[0.2.4] The position or token count at which the model behavior shifts from normal processing to truncation or error;

[0.2.5] The highest valid integer position index representable by the position_embedding scheme within context_length;

[0.2.6] The ratio of tokens actively processed to total available context_length; quantifies context_utilization as us