INDEX

Explanations

names with "Dr." prefix

oai_token-act-pair · gpt-3.5-turbo

references to the name "Dr." or related titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 homeland

-0.68

 exterior

-0.66

olen

-0.65

 bargaining

-0.64

 unsett

-0.63

 unpaid

-0.62

 kingdom

-0.62

 unit

-0.62

 bailed

-0.62

 sprint

-0.62

POSITIVE LOGITS

Dr

3.71

Dr

2.35

dr

2.05

Professor

1.90

Prof

1.65

Doctor

1.64

doctor

1.45

Doctors

1.43

DR

1.35

Mr

1.27

Activations Density 0.011%

names with "Dr." prefix

references to the name "Dr." or related titles

No Comments

No Known Activations