INDEX

Explanations

references to race-related terminology and legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 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

 impactful

-1.02

 microbiome

-1.00

 onboarding

-0.95

TLDR

-0.93

 curated

-0.92

qPCR

-0.91

 tasked

-0.90

 overarching

-0.88

 incentiv

-0.88

 showcasing

-0.85

POSITIVE LOGITS

 muß

1.47

 daß

1.47

 läßt

1.38

 müßte

1.26

 Moslem

1.25

 idéia

1.24

 Daß

1.18

 mußte

1.17

 mußten

1.10

faßt

1.08

Activations Density 4.704%

references to race-related terminology and legal contexts

No Comments

No Known Activations