INDEX

Explanations

mathematical expressions and calculations

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

Expressions with collected terms

np_acts-logits-general · gemini-2.0-flash

variables and numbers

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

Rom

-0.42

Ɓ

-0.40

 Trom

-0.38

 Román

-0.38

bcryptjs

-0.38

amo

-0.38

 gott

-0.37

ORROW

-0.37

 Romana

-0.37

 Mary

-0.36

POSITIVE LOGITS

 informée

0.55

rungsseite

0.53

 mijne

0.51

 miniaturka

0.50

 <=",

0.47

+#+#

0.47

tonode

0.46

 flecha

0.46

 hidrá

0.46

 '\\;'

0.45

Activations Density 1.238%

mathematical expressions and calculations

Expressions with collected terms

variables and numbers

No Comments

No Known Activations