INDEX

Explanations

clarity or emphasis in statements

oai_token-act-pair · gpt-3.5-turbo

phrases indicating clarity or making something clear

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

inse

-0.74

umbn

-0.68

umat

-0.66

asus

-0.65

ernels

-0.65

unte

-0.64

chance

-0.64

hovah

-0.63

Rou

-0.63

olen

-0.63

POSITIVE LOGITS

ances

0.91

 distinctions

0.89

 outlines

0.83

iary

0.81

 distinction

0.79

forth

0.74

 sailing

0.74

 clear

0.71

 contrasts

0.68

 clearer

0.68

Activations Density 0.019%

clarity or emphasis in statements

phrases indicating clarity or making something clear

No Comments

No Known Activations