INDEX

Explanations

references to various styles and formatting elements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

}))

-0.94

sizeCache

-0.92

存于互联网档案馆

-0.90

 immédi

-0.89

-0.88

 CanadaChoose

-0.87

Captor

-0.85

InSection

-0.85

 CSRF

-0.85

__':

-0.84

POSITIVE LOGITS

 styles

1.57

 Style

1.53

 STYLE

1.51

 Styles

1.50

 style

1.50

Style

1.46

STYLE

1.43

style

1.31

Styles

1.27

styles

1.26

Activations Density 0.032%

references to various styles and formatting elements

No Comments

No Known Activations