INDEX

Explanations

academic affiliations and contributions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

المناصب

-0.60

자인

-0.53

Eksteraj

-0.52

zkod

-0.51

 LCCN

-0.49

انجليز

-0.49

έν

-0.47

 pracovní

-0.46

ţ

-0.45

Caz

-0.45

POSITIVE LOGITS

 Korean

1.23

 Koreans

1.22

 korean

1.19

 Seoul

1.13

Korean

1.12

 Korea

1.11

 korea

1.10

Korea

1.09

Seoul

0.98

 kimchi

0.91

Activations Density 0.182%

academic affiliations and contributions

No Comments

No Known Activations