INDEX

Explanations

mentions of privacy and data handling practices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

EndInit

-0.67

ha

-0.62

Ha

-0.49

ModelForm

-0.49

板

-0.48

Ha

-0.47

skab

-0.47

 ModelRenderer

-0.46

lan

-0.45

liv

-0.45

POSITIVE LOGITS

 purpoſe

0.91

 iſt

0.82

 itſelf

0.81

Personendaten

0.77

 neceff

0.77

 Anſ

0.76

 reaſon

0.76

ſelves

0.75

 myſelf

0.73

ſelf

0.72

Activations Density 0.116%

mentions of privacy and data handling practices

No Comments

No Known Activations