INDEX

Explanations

automatic cardiac recorders

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

This neuron detects computer code and programming-related text, particularly focusing on code elements like function names, file paths, and numeric identifiers.

oai_token-act-pair · claude-3-7-sonnet-20250219 Triggered by @neilrathi

The neuron is activated by sequences of numeric and hexadecimal literals (e.g. addresses, opcode constants, and other hex‐style numbers) in low‐level code.

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

Not in Any Lists

No Comments

Negative Logits

 bebé

-1.49

ἵ

-1.43

υμ

-1.41

dası

-1.41

 çünkü

-1.38

 Offizielle

-1.38

langsung

-1.38

 skvě

-1.34

itaji

-1.34

seperti

-1.30

POSITIVE LOGITS

 portatil

1.60

 esquecer

1.59

but

1.48

lceil

1.44

 FAVORITE

1.38

ьез

1.30

rceil

1.30

 During

1.30

鏖

1.30

 However

1.29

Activations Density 0.014%

automatic cardiac recorders

This neuron detects computer code and programming-related text, particularly focusing on code elements like function names, file paths, and numeric identifiers.

The neuron is activated by sequences of numeric and hexadecimal literals (e.g. addresses, opcode constants, and other hex‐style numbers) in low‐level code.

No Comments

No Known Activations

automatic cardiac recorders

This neuron detects computer code and programming-related text, particularly focusing on code elements like function names, file paths, and numeric identifiers.

The neuron is activated by sequences of numeric and hexadecimal literals (e.g. addresses, opcode constants, and other hex‐style numbers) in low‐level code.

No Comments

No Known Activations