Bpe 8k
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
Bpe 8k has 16 facts recorded in Dontopedia across 8 references, with 3 live disagreements.
Mostly:has eos id(2), is tokenizer(2), has vocab size(2)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (10)
Other subjects in dontopedia point AT this entity as a value. These are inverse relationships — e.g. "X motherOf this subject" — and answer questions the forward facts can't. Grouped by predicate.
usesDatasetUses Dataset(2)
- Prototype Experiment
ex:prototype-experiment - Small Experiment
ex:small-experiment
believesFreqweightedHelpsBpe8kBelieves Freqweighted Helps Bpe8k(1)
- Xenonfun
ex:xenonfun
causesPoorCoverageForCauses Poor Coverage for(1)
- 200 Entry Limit
ex:200-entry-limit
hasSparseCacheForHas Sparse Cache for(1)
- Kickmodel
ex:kickmodel
helpsMoreForHelps More for(1)
- Freqweighted Eviction
ex:freqweighted-eviction
impliesBpe8kVocabImplies Bpe8k Vocab(1)
- Bpe Tokenizer
ex:bpe-tokenizer
includesVariantIncludes Variant(1)
- Tokenization Variants
ex:tokenization-variants
involvesInvolves(1)
- Task Run Bpe 8k
ex:task-run-bpe-8k
usesBPEUses Bpe(1)
- Optimizer Benchmark Gpt2 Scale
ex:optimizer-benchmark-gpt2-scale
Other facts (16)
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 |
|---|---|---|
| Has Eos Id | 1 | [1] |
| Has Eos Id | 1 | [2] |
| Is Tokenizer | true | [1] |
| Is Tokenizer | BPE-8K | [2] |
| Has Vocab Size | 8192 | [1] |
| Has Vocab Size | 8000 | [3] |
| Rdf:type | Dataset | [7] |
| Rdf:type | Tokenizer | [8] |
| Has Vocab | 8192 | [2] |
| Requires Minimum Cache Entries | 1000 | [3] |
| Has Cache Coverage | 2.5% | [3] |
| Includes All Improvements | Report | [4] |
| Should Run With Improvements | Report | [4] |
| Works End to End | true | [5] |
| Currently Sufficient | true | [5] |
| Presupposes Tokenization | null | [6] |
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 (8)
ctx:discord/blah/watt-activation/part-232ctx:discord/blah/watt-activation/part-248ctx:discord/blah/watt-activation/part-480ctx:discord/blah/watt-activation/part-481ctx:discord/blah/watt-activation/part-485ctx:discord/blah/watt-activation/part-483ctx:discord/blah/watt-activation/275- full textwatt-activation-275text/plain3 KB
doc:agent/watt-activation-275/149bf24b-54fb-4412-b289-e2e03bccffe2Show excerpt
[2026-03-13 20:46] xenonfun: ⏺ This tells a clear story: Current trained weights are NOT naturally ternary. Only ~46% of values are near {-1, 0, 1} after scaling (random Gaussian would give ~47%, so no better than chance). Relative err…
ctx:discord/blah/watt-activation/479- full textwatt-activation-479text/plain2 KB
doc:agent/watt-activation-479/cf877f60-5f22-46ef-a130-a278610bc58dShow excerpt
[2026-03-21 23:05] xenonfun: ``` ⏺ All committed and pushed. Server is live at http://localhost:42069/ with full controls. Final session stats: ┌────────────────────┬──────────────────┬───────────────────────────────┐ │ Metric…
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.