INDEX

Explanations

references to works and authors in various academic and literary contexts

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

 للمعارف

-0.49

 üzere

-0.45

 AspNetCore

-0.45

 numele

-0.44

washingtonpost

-0.42

“

-0.42

वा

-0.42

回る

-0.42

 luego

-0.42

 perhatian

-0.41

POSITIVE LOGITS

 guide

1.58

 Guide

1.43

guide

1.38

Guide

1.34

 handbook

1.32

 GUIDE

1.27

 guides

1.20

 Handbook

1.19

GUIDE

1.16

 Guida

1.09

Activations Density 0.253%

references to works and authors in various academic and literary contexts

No Comments

No Known Activations