INDEX

Explanations

names of political figures

oai_token-act-pair · gpt-3.5-turbo

proper nouns and names associated with significant figures and events

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

ARDIS

-0.81

emort

-0.78

âĸĪâĸĪâĸĪâĸĪ

-0.71

 Myster

-0.70

 merc

-0.69

 Luthor

-0.68

sylv

-0.67

Newsletter

-0.66

EXT

-0.64

Cub

-0.63

POSITIVE LOGITS

ji

1.05

 Sharma

1.04

bh

1.01

 Singh

1.01

 Sabha

1.00

 Bhar

0.98

ya

0.97

 Patel

0.95

umbai

0.95

Raj

0.94

Activations Density 0.333%

names of political figures

proper nouns and names associated with significant figures and events

No Comments

No Known Activations