INDEX

Explanations

references to images or visual content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

UnusedPrivate

-0.86

@"/

-0.83

/>";

-0.81

providedIn

-0.71

 traditionnels

-0.69

 paroisse

-0.68

RegressionTest

-0.66

 przyzna

-0.66

IsPostBack

-0.66

troppo

-0.65

POSITIVE LOGITS

 image

2.42

 images

2.38

 Image

2.28

image

2.12

 Images

2.10

 IMAGE

2.07

images

2.01

Image

2.00

IMAGE

1.92

Images

1.87

Activations Density 0.055%

references to images or visual content

No Comments

No Known Activations