INDEX

Explanations

It seems to be activating on mathematical expressions (especially those with symbols like '&' and '$'), code snippets in HTML format, and numbers separated by commas.

oai_token-act-pair · gemini-2.0-flash

code-like structures and symbols

np_token-act-pair-logits · gpt-4o-mini

Code and math expressions

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nalités

-0.65

RenderAtEndOf

-0.63

!*\

-0.59

ритори

-0.57

Général

-0.54

__':

-0.53

 XCTest

-0.53

AsUp

-0.53

setLayoutManager

-0.51

siąż

-0.51

POSITIVE LOGITS

Vorlage

0.51

withOpacity

0.51

IBOutlet

0.50

cule

0.49

 point

0.46

点的

0.45

FlatStyle

0.44

źni

0.44

 Brem

0.44

 видом

0.44

Activations Density 0.062%

It seems to be activating on mathematical expressions (especially those with symbols like '&' and '$'), code snippets in HTML format, and numbers separated by commas.

code-like structures and symbols

Code and math expressions

No Comments

No Known Activations

It seems to be activating on mathematical expressions (especially those with symbols like '&' and '$'), code snippets in HTML format, and numbers separated by commas.

code-like structures and symbols

Code and math expressions

No Comments

No Known Activations