INDEX

Explanations

references to particular datasets or indexes associated with research protocols

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

 Lucre

-0.56

 Alejandra

-0.55

 Estelle

-0.51

kwe

-0.49

 Gower

-0.47

 Dawes

-0.46

θρώ

-0.46

 Francesca

-0.46

 Faure

-0.46

PSC

-0.45

POSITIVE LOGITS

CI

1.07

CI

1.06

PI

1.04

PI

1.02

SI

1.00

EI

0.99

MI

0.98

GI

0.97

EI

0.96

TI

0.96

Activations Density 0.516%

references to particular datasets or indexes associated with research protocols

No Comments

No Known Activations