INDEX

Explanations

scientific terms and identifiers related to entities, particularly in a technical or factual context

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

zinger

-0.57

ativement

-0.56

 presidency

-0.54

 Davison

-0.52

 Somerville

-0.52

 vesti

-0.52

ubscribe

-0.50

 Dickey

-0.50

 disponibilités

-0.49

bergen

-0.49

POSITIVE LOGITS

Dea

0.71

 coar

0.70

rea

0.65

 Mead

0.65

IAA

0.64

Ea

0.63

 aapt

0.63

LookAnd

0.62

EAP

0.62

EEA

0.62

Activations Density 0.358%

scientific terms and identifiers related to entities, particularly in a technical or factual context

No Comments

No Known Activations