INDEX

Explanations

mentions of high-ranking positions or titles within a company

oai_token-act-pair · gpt-3.5-turbo

references to individuals in corporate or professional roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

riad

-0.66

court

-0.63

 nurs

-0.61

obyl

-0.60

 mentally

-0.57

 extrem

-0.57

banks

-0.56

 jaws

-0.55

 insurers

-0.55

 magnitude

-0.54

POSITIVE LOGITS

VP

1.28

olicy

0.88

ledged

0.81

/-

0.80

osate

0.77

PO

0.76

uably

0.76

retty

0.76

OTUS

0.75

YL

0.75

Activations Density 0.003%

mentions of high-ranking positions or titles within a company

references to individuals in corporate or professional roles

No Comments

No Known Activations