INDEX

Explanations

mentions of formal discussions or panel events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

indd

-0.63

 useDispatch

-0.54

Frequently

-0.51

JFrame

-0.51

 hereof

-0.49

Hauptartikel

-0.49

 CELEBR

-0.48

Thousands

-0.48

-0.47

 Thousands

-0.46

POSITIVE LOGITS

 motion

0.81

 motions

0.70

 remarks

0.69

 presenters

0.68

 minutes

0.68

 introductions

0.67

 adjourn

0.66

 adjournment

0.66

RegressionTest

0.65

 briefly

0.65

Activations Density 0.218%

mentions of formal discussions or panel events

No Comments

No Known Activations