INDEX

Explanations

words related to specific professional roles or organizations

oai_token-act-pair · gpt-3.5-turbo

nouns related to professions and roles in society

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

CHAPTER

-0.72

FINE

-0.70

amina

-0.66

 meas

-0.63

 pleas

-0.62

ormonal

-0.62

 corrid

-0.61

 fortun

-0.60

haar

-0.60

abus

-0.59

POSITIVE LOGITS

iasco

0.72

 Crimes

0.70

Cyn

0.69

LLP

0.68

utra

0.65

plates

0.64

Ltd

0.63

 Drops

0.62

plex

0.62

 Square

0.62

Activations Density 0.605%

words related to specific professional roles or organizations

nouns related to professions and roles in society

No Comments

No Known Activations