INDEX

Explanations

references to significant sizes or magnitudes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/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.21.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

ständig

-0.42

ESTAMP

-0.42

maktadır

-0.40

 pełni

-0.39

 presso

-0.39

zde

-0.39

izoen

-0.38

하였다

-0.37

unakan

-0.37

 Hause

-0.37

POSITIVE LOGITS

BIG

0.97

big

0.95

BIG

0.93

big

0.93

Big

0.90

bigger

0.85

 bigger

0.82

Big

0.80

Biggest

0.78

biggest

0.77

Activations Density 0.118%

references to significant sizes or magnitudes

No Comments

No Known Activations