INDEX

Explanations

names or components of names related to persons or entities

oai_token-act-pair · gpt-3.5-turbo

names of individuals or entities involved in a situation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 enthusi

-0.92

 exting

-0.90

 pione

-0.84

 eleph

-0.81

ò

-0.73

é¾įå¥ĳå£«

-0.69

 tremend

-0.68

DeliveryDate

-0.67

AFL

-0.67

ÃĥÃĤÃĥÃĤ

-0.66

POSITIVE LOGITS

oud

0.97

asy

0.95

ap

0.89

imov

0.89

houn

0.89

ibi

0.89

ient

0.88

arnaev

0.88

emed

0.86

ipal

0.86

Activations Density 0.168%

names or components of names related to persons or entities

names of individuals or entities involved in a situation

No Comments

No Known Activations