INDEX

Explanations

Code/math/strings

np_max-act · gemini-2.0-flash

This neuron fires on floating-point numeric literals (decimal numbers) in the text.

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

snippets of source code (programming tokens and identifiers).

oai_token-act-pair · gpt-5-mini Triggered by @vetterc0

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

那样

-0.06

 Bowl

-0.06

 Subscription

-0.06

 Shepard

-0.06

 getTotal

-0.06

ichtig

-0.06

rim

-0.06

='#

-0.06

 chair

-0.06

 Super

-0.06

POSITIVE LOGITS

mov

0.07

 LOGGER

0.07

↵

0.06

充

0.06

ournemouth

0.06

्यप

0.06

(++

0.06

 clandest

0.06

Submitted

0.06

CLUD

0.06

Activations Density 0.425%

Code/math/strings

This neuron fires on floating-point numeric literals (decimal numbers) in the text.

snippets of source code (programming tokens and identifiers).

No Comments

No Known Activations