Last Hidden State Extraction
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-10.)
Last Hidden State Extraction has 5 facts recorded in Dontopedia across 2 references.
Mostly:uses slicing(1), calls detach(1), calls numpy(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedOther facts (5)
The long tail: predicates that appear too rarely to warrant their own section. Filter or scroll to find a specific one. Each row links to its source.
| Predicate | Value | Ref |
|---|---|---|
| Uses Slicing | [:, 0, :] | [1] |
| Calls Detach | Detach | [1] |
| Calls Numpy | Numpy | [1] |
| Rdf:type | Tensor Slicing | [2] |
| Extracts | First Token | [2] |
Timeline
Timeline axis is valid_time — when each source says the fact was true in the world, not when Dontopedia learned about it. Retracted rows are kept for provenance; coloured stripes indicate the context kind.
References (2)
ctx:claims/beam/56b422f7-45b6-49d7-9022-6df268bf77c3- full textbeam-chunktext/plain1 KB
doc:beam/56b422f7-45b6-49d7-9022-6df268bf77c3Show excerpt
inputs = tokenizer(document, return_tensors='pt') outputs = model(**inputs) return outputs.last_hidden_state[:, 0, :].detach().numpy() # vectorize 10K documents documents = [...] # list of 10K documents vectors = [vectorize_do…
ctx:claims/beam/24776806-43b0-491e-806d-e4f4e8d75851
See also
Keep researching
Missing something or suspicious of what's here? Kick off a research session — a Claude agent will investigate, cite its sources, and file new facts into a dedicated context you can review before accepting into the shared view.