Unverified50% confidenceFactExact time

The GPT-4o tokenizer turns 'How many r's are in strawberry?' into 8 tokens: How | ·many | ·r | 's | ·are | ·in | ·strawberry | ?, where 'strawberry' is token #101830.

1

Sources

50%

Confidence

Medium-term (~90 days)

Relevance

10/13/2026

First Seen

Valid until: 1/11/2027

Sources

Why Can't LLMs Count the R's in Strawberry? A Deep Dive into Tokenizer Mechanics

Related Entities

Related Claims

UnverifiedEncoded with o200k_base, 'I love strawberry' becomes [40, 3047, 101830].63% similar UnverifiedThe miscounting of the character 'r' stems from the tokenizer splitting 'strawberry' into a single token, so the model cannot directly access its internal character composition.62% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/1022420

API

curl https://kongchang.com/api/v1/knowledge/claims/1022420

MCP

get_claim(id=1022420)