INDEX

Explanations

names or words containing "ib"

oai_token-act-pair · gpt-3.5-turbo

references to a specific individual, likely a historian or writer

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

 Ceres

-0.67

 Blizzard

-0.67

 Samoa

-0.66

 Russ

-0.65

ISTER

-0.64

 marsh

-0.63

isters

-0.62

CPS

-0.62

 Irma

-0.62

 femin

-0.60

POSITIVE LOGITS

ilib

1.20

raltar

1.05

ibl

1.03

bole

0.99

ulous

0.97

odies

0.96

rahim

0.92

uilt

0.92

acter

0.92

iotics

0.91

Activations Density 0.013%

names or words containing "ib"

references to a specific individual, likely a historian or writer

No Comments

No Known Activations