INDEX

Explanations

numbers and their formatting, likely in a programming or data context

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

CardModule

-0.34

 مس

-0.33

 кло

-0.29

is

-0.28

-0.26

 Оно

-0.26

innis

-0.25

ipedi

-0.25

POSITIVE LOGITS

httphttps

0.85

Personendaten

0.74

 Comprometido

0.72

 zwiſchen

0.69

Diwedd

0.69

 informée

0.68

pecabe

0.67

ConstraintMaker

0.66

 Geiſt

0.65

 Meksiku

0.65

Activations Density 1.236%

numbers and their formatting, likely in a programming or data context

No Comments

No Known Activations