INDEX

Explanations

phrases expressing personal needs or desires

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 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

}{*}{

-0.54

 most

-0.47

bah

-0.45

;'>

-0.44

ății

-0.44

bano

-0.43

ījum

-0.42

also

-0.42

geführt

-0.42

اه

-0.42

POSITIVE LOGITS

 propOrder

0.95

 sundry

0.80

 OFDb

0.77

ArrowToggle

0.75

UserScript

0.73

providedIn

0.71

adaptiveStyles

0.69

 Paglinawan

0.69

ſelf

0.69

ualaikum

0.68

Activations Density 0.087%

phrases expressing personal needs or desires

No Comments

No Known Activations