INDEX

Explanations

the word "cle" followed by a number indicating a rating or position

oai_token-act-pair · gpt-3.5-turbo

words that begin with "cle" and their variations

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

 Ducks

-0.75

yrinth

-0.75

 Krug

-0.70

 Galaxy

-0.64

 Trafford

-0.64

BuyableInstoreAndOnline

-0.64

 Kings

-0.63

UME

-0.62

 Shogun

-0.62

Tul

-0.61

POSITIVE LOGITS

arest

1.14

avage

1.02

veland

0.95

combe

0.90

asp

0.84

utic

0.83

cle

0.82

ipe

0.82

leans

0.81

llan

0.81

Activations Density 0.004%

the word "cle" followed by a number indicating a rating or position

words that begin with "cle" and their variations

No Comments

No Known Activations