INDEX

Explanations

questions related to preferences and opinions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

-0.48

nonumber

-0.46

su

-0.45

pas

-0.45

tagne

-0.43

ฒ

-0.43

須

-0.42

 manufacturer

-0.41

</i>

-0.41

 lavander

-0.40

POSITIVE LOGITS

GEBURTSDATUM

0.99

 betweenstory

0.86

dafx

0.78

 autorytatywna

0.76

ScopeManager

0.75

Билгалдахарш

0.69

 tfsi

0.68

 photolibrary

0.68

writeFieldEnd

0.67

 jsPsych

0.67

Activations Density 0.003%

questions related to preferences and opinions

No Comments

No Known Activations