INDEX

Explanations

directional or positional phrases that suggest movement or escape

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 alſo

-0.52

 juſt

-0.52

 uſed

-0.51

 firſt

-0.51

 auffi

-0.51

 ſur

-0.47

 faſt

-0.47

 whoſe

-0.47

 chiaro

-0.47

Eſ

-0.45

POSITIVE LOGITS

 يتيمه

0.84

LookAnd

0.76



0.74

 autorytatywna

0.73

migrationBuilder

0.70

 kasarigan

0.70

 Biôgrafia

0.70

存于互联网档案馆

0.69

帖最后由

0.68

Jeografia

0.66

Activations Density 0.162%

directional or positional phrases that suggest movement or escape

No Comments

No Known Activations