INDEX

Explanations

names or acronyms related to medical or health-related organizations

oai_token-act-pair · gpt-3.5-turbo

references to medical and health-related topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

lihood

-0.76

 ç¥ŀ

-0.75

eers

-0.74

esville

-0.71

glers

-0.71

 Corpus

-0.67

creen

-0.67

terson

-0.66

opol

-0.66

 Samoa

-0.64

POSITIVE LOGITS

senal

1.46

abbit

1.08

utherford

0.98

VP

0.90

ULE

0.89

ADE

0.89

ACK

0.89

Ds

0.89

ansom

0.88

RR

0.85

Activations Density 0.072%

names or acronyms related to medical or health-related organizations

references to medical and health-related topics

No Comments

No Known Activations