INDEX

Explanations

leader

np_max-act · gemini-2.0-flash

The neuron selectively fires on technical or quantitative jargon—terms you’d see in programming error messages, math/stats discussions, or other technical contexts.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 democracy

-0.07

 championships

-0.07

ріб

-0.07

 ovšem

-0.07

ován

-0.06

 Willie

-0.06

Tek

-0.06

aoke

-0.06

ея

-0.06

 erupted

-0.06

POSITIVE LOGITS

�

0.07

/antlr

0.06

 propelled

0.06

व

0.06

(boolean

0.06

ButtonTitles

0.06

aget

0.06

modifiers

0.06

 фут

0.06

ंय

0.05

Activations Density 0.044%

leader

The neuron selectively fires on technical or quantitative jargon—terms you’d see in programming error messages, math/stats discussions, or other technical contexts.

No Comments

No Known Activations