INDEX

Explanations

references to companies, organizations, and legal entities

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

 propOrder

-1.10

)");

-1.06

 Theſe

-1.00

ſelves

-0.97

 itſelf

-0.97

 Monfieur

-0.95

ReusableCell

-0.94

 ―――――

-0.94

 greateſt

-0.94

 Jefus

-0.93

POSITIVE LOGITS

Inc

2.13

LLC

1.75

Ltd

1.55

Inc

1.48

INC

1.36

inc

1.35

inc

1.21

 Incorporated

1.20

 Corp

1.15

 GmbH

1.11

Activations Density 0.396%

references to companies, organizations, and legal entities

No Comments

No Known Activations