INDEX

Explanations

references to the letter 'B' or its variations in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ulose

-0.51

מבר

-0.50

zusetzen

-0.50

ukų

-0.50

 myſelf

-0.48

ecap

-0.47

 Laub

-0.47

出版年

-0.47

 contextLoads

-0.47

 Вікіпе

-0.47

POSITIVE LOGITS

 nakalista

0.64

 tématu

0.59

metrist

0.57

 bVar

0.57

 kimono

0.56

eun

0.56

elemField

0.54

 doInBackground

0.54

 бумаги

0.54

prelude

0.54

Activations Density 0.294%

references to the letter 'B' or its variations in the text

No Comments

No Known Activations