INDEX

Explanations

words related to enabling or allowing something to happen

oai_token-act-pair · gpt-3.5-turbo

phrases and variations of the word "enable" indicating empowerment or facilitation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

boy

-0.84

bell

-0.82

alian

-0.80

worn

-0.77

ynski

-0.71

asso

-0.69

been

-0.68

iform

-0.68

bas

-0.67

ĳå£«

-0.67

POSITIVE LOGITS

us

0.86

 interoper

0.77

ment

0.76

 access

0.75

ezvous

0.73

jriwal

0.73

 speedy

0.72

ments

0.72

 seamless

0.72

 penetration

0.72

Activations Density 0.043%

words related to enabling or allowing something to happen

phrases and variations of the word "enable" indicating empowerment or facilitation

No Comments

No Known Activations