INDEX

Explanations

wrestling promo or script

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

The neuron lights up strongly on pieces of text that repeat verbatim—i.e. it flags duplicate or exactly repeated spans in the document.

oai_token-act-pair · o4-mini Triggered by @jyhe0408

repetitive or duplicated text patterns within documents.

oai_token-act-pair · claude-4-5-sonnet Triggered by @jyhe0408

phrases from practical, procedural, or list-like content that mention concrete items or steps (e.g., products, directions, rules, or menu entries).

oai_token-act-pair · gpt-5 Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ْنَ

0.76

sthe

0.73

спекти

0.72

nen

0.71

্স

0.71

lage

0.70

ts

0.68

UIImage

0.68

ls

0.68

prong

0.66

POSITIVE LOGITS

 amaç

0.83

 Cuthbert

0.77

 polypropylene

0.76

的话

0.74

烂

0.74

内置

0.72

찾

0.72

 rattan

0.71

愕

0.71

不能

0.70

Activations Density 0.000%

wrestling promo or script

The neuron lights up strongly on pieces of text that repeat verbatim—i.e. it flags duplicate or exactly repeated spans in the document.

repetitive or duplicated text patterns within documents.

phrases from practical, procedural, or list-like content that mention concrete items or steps (e.g., products, directions, rules, or menu entries).

No Comments

No Known Activations

wrestling promo or script

The neuron lights up strongly on pieces of text that repeat verbatim—i.e. it flags duplicate or exactly repeated spans in the document.

repetitive or duplicated text patterns within documents.

phrases from practical, procedural, or list-like content that mention concrete items or steps (e.g., products, directions, rules, or menu entries).

No Comments

No Known Activations