INDEX

Explanations

instances of the word "introduce" and its variants, indicating a focus on introducing concepts or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 dezelve

-0.54

 pikir

-0.49

ciphertext

-0.48

 gempa

-0.45

 Waller

-0.45

 alqu

-0.44

 zoude

-0.43

inerja

-0.43

 inoxydable

-0.42

 regalías

-0.42

POSITIVE LOGITS

 introduction

1.23

 introduces

1.20

 introduce

1.19

 introductions

1.16

 Introduce

1.11

 introduced

1.11

introduce

1.10

Introduce

1.05

 Introdu

1.05

introdu

1.01

Activations Density 0.111%

instances of the word "introduce" and its variants, indicating a focus on introducing concepts or entities

No Comments

No Known Activations