INDEX

Explanations

phrases indicating the importance or relevance of certain concepts

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

 Cæsar

-0.64

คุณ

-0.59

 invitamos

-0.57

 يتيمه

-0.57

//

-0.53

TypedDataSet

-0.51

 laikā

-0.50

 ferons

-0.50

 nifer

-0.49

argout

-0.49

POSITIVE LOGITS

 course

1.57

course

1.41

COURSE

1.18

 sorts

1.15

 COURSE

1.11

 Course

0.99

Course

0.95

the

0.94

sted

0.91

 yore

0.84

Activations Density 1.631%

phrases indicating the importance or relevance of certain concepts

No Comments

No Known Activations