INDEX

Explanations

Immunohistochemical, Q:

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

code-related content including programming syntax, HTML, CSS, and technical documentation.

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

Triggered by @neilrathi

The neuron strongly activates on programming‐language and markup tokens (e.g. code keywords, symbols, HTML/CSS/JS identifiers), effectively detecting code snippets.

oai_token-act-pair · o4-mini

Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_34/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 પર

-0.79

Vom

-0.79

静

-0.76

EINVAL

-0.73

 Walgreens

-0.69

财务

-0.69

ツール

-0.68

機関

-0.68

σει

-0.68

Rü

-0.68

POSITIVE LOGITS

rscheinlichkeit

0.87

ška

0.87

้ม

0.82

 spectrometry

0.81

pective

0.79

DCC

0.79

elt

0.78

ল

0.77

傷害

0.77

NOC

0.76

Activations Density 0.035%

Immunohistochemical, Q:

code-related content including programming syntax, HTML, CSS, and technical documentation.

The neuron strongly activates on programming‐language and markup tokens (e.g. code keywords, symbols, HTML/CSS/JS identifiers), effectively detecting code snippets.

No Known Activations

Immunohistochemical, Q:

code-related content including programming syntax, HTML, CSS, and technical documentation.

The neuron strongly activates on programming‐language and markup tokens (e.g. code keywords, symbols, HTML/CSS/JS identifiers), effectively detecting code snippets.

No Known Activations