INDEX

Explanations

phrases related to automated processes or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

Portale

-0.73

Datuak

-0.71

PasswordEncoder

-0.69

تقاوى

-0.68

Personensuche

-0.68

parency

-0.66

 ciudadana

-0.65

bkz

-0.64

initializeApp

-0.64

 виправивши

-0.64

POSITIVE LOGITS

 Auto

1.31

 auto

1.24

Auto

1.20

auto

1.11

AUTO

1.06

 AUTO

1.02

 авто

0.82

 phase

0.71

 automatic

0.69

uto

0.69

Activations Density 0.156%

phrases related to automated processes or actions

No Comments

No Known Activations