INDEX

Explanations

references to murders and related circumstances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 AssemblyCulture

-0.59

 giras

-0.32

 average

-0.32

 erity

-0.31

 ligera

-0.29

 meaning

-0.29

 trous

-0.29

 tables

-0.28

 performance

-0.28

 large

-0.28

POSITIVE LOGITS

RegressionTest

0.79

setVerticalGroup

0.64

WEBPACK

0.62

CloseOperation

0.60

undai

0.59

AsUp

0.58

hower

0.57

ImageContext

0.56

 murdered

0.56

murder

0.56

Activations Density 0.455%

references to murders and related circumstances

No Comments

No Known Activations