INDEX

Explanations

references to authority figures, particularly those denoted as "Lord."

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

XS

-0.64

tagHelperRunner

-0.62

︎

-0.62

:],

-0.61

脚注の使い方

-0.60

 Atra

-0.60

 selections

-0.58

փ

-0.58

seud

-0.57

Atra

-0.57

POSITIVE LOGITS

 Lord

3.22

Lord

2.96

 lord

2.79

 LORD

2.73

LORD

2.40

 Lords

2.30

lord

2.24

 lords

2.17

 Seigneur

1.77

lords

1.61

Activations Density 0.065%

references to authority figures, particularly those denoted as "Lord."

No Comments

No Known Activations