Byte-pair encoding
Algorithm for data compression and tokenization that iteratively replaces the most frequent pairs of consecutive bytes or symbols with a new representation
Nº Q860962 ★
Common · Knowledge
Byte-pair encoding
Algorithm for data compression and tokenization that iteratively replaces the most frequent pairs of consecutive bytes or symbols with a new representation
In computing, byte-pair encoding (BPE), or digram coding, is an algorithm, first described in 1994 by Philip Gage, for encoding strings of text into smaller strings by creating and using a translation table. A slightly modified version of the algorithm is used in large language model tokenizers.
Last price
—
Floor price
—
7-day median
—
30-day sales
0
30-day range
—
In circulation
0
Price history
median
low – high
sales
No sales in this period
Show table
| Date | median | Low | High | sales |
|---|
Sales history
- Last sale
- —
- 30-day average
- —
- 30-day low
- —
- 30-day high
- —
- Sales 7d
- 0
- Sales 30d
- 0
No sales yet.
Anonymous sales: no buyer or seller shown. Figures count player-to-player sales only.
From Wikipedia
In computing, byte-pair encoding (BPE), or digram coding, is an algorithm, first described in 1994 by Philip Gage, for encoding strings of text into smaller strings by creating and using a translation table. A slightly modified version of the algorithm is used in large language model tokenizers. The original version of the algorithm focused on compression. It replaces the highest-frequency pair of bytes with a new byte that was not contained in the initial dataset. A lookup table of the replacements is required to rebuild the initial dataset. The modified version builds "tokens" (units of recognition) that match varying amounts of source text, from single characters (including single digits or single punctuation marks) to whole words (even long compound words).
Text: Wikipédia, CC BY-SA 4.0. ·
