I was thinking about LLM tokenization (as one does) and had a thought: We select the next output token for an LLM based on its likelihood, but shorter tokens are more likely.

Why? Shorter common tokens are (correctly) learned to be higher-probability because they have the combined probability of any word they could complete. However, standard generation techniques will only consider a subset of probabilities and scale the largest probabilities. Both of these will take the highest probabilities and increase them further, meaning short/common tokens become significantly more likely to be generated just because they're shorter.

I ran an experiment to investigate this, showing that the first-character distribution of words generated by nanoGPT[1] is similar regardless of tokenization without top-K or temperature scaling, but if we use common settings (top-K=200 and temperature=0.8), we can increase the likelihood that a word starts with 'c' from 4% up to 10% just [...]

---

Outline:

(01:36) Why?

(02:27) Top-K Sampling

(02:52) Temperature 1.0

(04:19) The Experiment

(05:42) Results

(06:47) Shakespeare Experiment

(07:08) Results

(08:15) Why Does It Matter?

The original text contained 4 footnotes which were omitted from this narration.

---

First published:
August 24th, 2025

Source:
https://www.lesswrong.com/posts/iZPKuuWsDXAcQWbLJ/shorter-tokens-are-more-likely

---

Narrated by TYPE III AUDIO.

---

Images from the article:

Probability tree diagram showing branching paths with percentage splits and outcomes.

Probability tree diagram showing branching paths with percentage distributions.

Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

“Shorter Tokens Are More Likely” by Brendan Long

LessWrong (30+ Karma)

“Shorter Tokens Are More Likely” by Brendan Long