INDEX

Explanations

words starting with the sequence "wh"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

findpost

-0.55

MLLoader

-0.54

setupUi

-0.53

ंदीखरीदारी

-0.49

addContainerGap

-0.49

 مشين

-0.48

 IBOutlet

-0.47

InSection

-0.47

eye

-0.47

StoryboardSegue

-0.47

POSITIVE LOGITS

wh

1.05

wh

0.97

Wh

0.93

Wh

0.91

ub

0.71

WH

0.69

WH

0.68

 whale

0.59

Whale

0.57

pub

0.57

Activations Density 1.093%

words starting with the sequence "wh"

No Comments

No Known Activations