INDEX

Explanations

terms related to specific individuals and their purposes or roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/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.0.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

 overall

-0.29

 nearly

-0.29

 смер

-0.29

 المعيارى

-0.29

 interested

-0.29

 features

-0.28

 âgé

-0.28

 highly

-0.28

 Bolívar

-0.28

 zostało

-0.28

POSITIVE LOGITS

Демографія

0.71

 ligiloj

0.70

PREFERRED

0.70

 esternos

0.69

enetrable

0.68

 Chwiliwch

0.66

 Pass

0.66

ConstraintMaker

0.65

utives

0.65

alyptus

0.64

Activations Density 0.210%

terms related to specific individuals and their purposes or roles

No Comments

No Known Activations