INDEX

Explanations

references to dates and event occurrences

oai_token-act-pair · gpt-3.5-turbo

punctuations and numerical representations within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 withd

-0.81

utherland

-0.79

anyl

-0.76

 mosqu

-0.71

wcs

-0.69

phia

-0.67

rimp

-0.65

blers

-0.62

uron

-0.62

alysed

-0.61

POSITIVE LOGITS

[+

0.78

assetsadobe

0.72

 Introduction

0.70

 COUR

0.70

 Retrieved

0.68

 Compl

0.68

ATT

0.68

 Conclusion

0.67

 Mutant

0.67

 Intro

0.66

Activations Density 0.119%

references to dates and event occurrences

punctuations and numerical representations within the text

No Comments

No Known Activations