INDEX

Explanations

references to children or kids

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

ducer

-0.81

AndEndTag

-0.79

 Astra

-0.76

Councillor

-0.74

trata

-0.73

 Hara

-0.72

 Tomlinson

-0.72

}';

-0.72

)";

-0.72

alena

-0.71

POSITIVE LOGITS

 kids

1.59

kids

1.41

kid

1.32

 Kids

1.30

Kids

1.26

 KIDS

1.25

KID

1.10

Kid

1.04

Kid

1.03

 guys

1.00

Activations Density 0.054%

references to children or kids

No Comments

No Known Activations