INDEX

Explanations

words related to medical or health issues, such as "breach" and "surgery"

oai_token-act-pair · gpt-3.5-turbo

words related to the concept of "break" or "broken."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ãħĭ

-0.85

å°Ĩ

-0.76

iating

-0.75

ulp

-0.73

 hemor

-0.71

 crumbling

-0.69

urated

-0.69

uyomi

-0.69

iates

-0.68

ournal

-0.67

POSITIVE LOGITS

bre

1.19

itbart

1.13

vity

1.03

cht

1.00

tto

0.98

lla

0.98

aches

0.97

ake

0.92

tt

0.91

akers

0.90

Activations Density 0.007%

words related to medical or health issues, such as "breach" and "surgery"

words related to the concept of "break" or "broken."

No Comments

No Known Activations