INDEX

Explanations

references to people, specifically names or titles associated with individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

RectangleBorder

-0.91

ientôt

-0.88

 JpaRepository

-0.76

LayoutConstraint

-0.70

setVerticalGroup

-0.70

ftagPool

-0.70

 disambiguazione

-0.68

IsContent

-0.66

 AssemblyVersion

-0.64

quiagem

-0.63

POSITIVE LOGITS

Mr

0.82

Mr

0.74

 AppColors

0.64

Mrs

0.64

 Herrn

0.63

 délé

0.63

 monsieur

0.62

Ms

0.61

 iVar

0.61

pyx

0.59

Activations Density 0.085%

references to people, specifically names or titles associated with individuals

No Comments

No Known Activations