INDEX

Explanations

fossils

np_acts-logits-general · gemini-2.5-flash-lite

The neuron activates for abbreviations and words starting with "Fo" like "Fossil," "Fossils," "Follwo," and "Foe."

oai_token-act-pair · claude-3-7-sonnet-20250219

Triggered by @neilrathi

The neuron fires on the two‐letter sequence “fo” (in any casing) at the start of tokens.

oai_token-act-pair · o4-mini

Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_10/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

↵

-2.72

 oculta

-2.45

which

-2.44

 verminder

-2.39

 kılıf

-2.31

 egne

-2.17

 teens

-2.16

ization

-2.13

 appropri

-2.11

ﺲ

-2.11

POSITIVE LOGITS

ropractor

2.61

fo

2.59

 για

2.56

2.39

?"

2.34

谢谢你

2.33

[]>(

2.31

replyreport

2.30

 другим

2.27

苼

2.27

Activations Density 0.014%

fossils

The neuron activates for abbreviations and words starting with "Fo" like "Fossil," "Fossils," "Follwo," and "Foe."

The neuron fires on the two‐letter sequence “fo” (in any casing) at the start of tokens.

No Known Activations

fossils

The neuron activates for abbreviations and words starting with "Fo" like "Fossil," "Fossils," "Follwo," and "Foe."

The neuron fires on the two‐letter sequence “fo” (in any casing) at the start of tokens.

No Known Activations