INDEX

Explanations

special formatting or symbolic representations in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

-0.61

tania

-0.56

hal

-0.53

«

-0.52

-0.51

ρης

-0.51

hout

-0.49

 stara

-0.47

baye

-0.47

POSITIVE LOGITS

 itſelf

1.07

1.02

 للمعارف

0.92

 betweenstory

0.92

 neceff

0.89

 AssemblyCulture

0.88

 Reſ

0.87

 Majefty

0.86

 himſelf

0.86

Efq

0.84

Activations Density 0.166%

special formatting or symbolic representations in the text

No Comments

No Known Activations