INDEX

Explanations

people in leadership positions, particularly executive directors

oai_token-act-pair · gpt-3.5-turbo

titles and positions of authority figures, specifically directors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

endish

-0.72

 favour

-0.68

tnc

-0.66

burning

-0.65

asca

-0.65

 rive

-0.64

nikov

-0.63

ulz

-0.62

eros

-0.61

illet

-0.60

POSITIVE LOGITS

Publisher

0.77

ially

0.76

 extraord

0.73

Anne

0.72

IAL

0.72

 Shaun

0.71

 Betsy

0.70

hips

0.70

Russ

0.70

 Stacy

0.68

Activations Density 0.124%

people in leadership positions, particularly executive directors

titles and positions of authority figures, specifically directors

No Comments

No Known Activations