INDEX

Explanations

phrases related to viewpoints or perspectives

oai_token-act-pair · gpt-3.5-turbo

references to different perspectives or viewpoints

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

active

-0.86

utenberg

-0.83

usable

-0.81

hai

-0.79

zai

-0.79

kell

-0.74

netflix

-0.74

iggs

-0.73

wagen

-0.72

Ku

-0.70

POSITIVE LOGITS

 mater

0.78

 IMAGES

0.78

enance

0.73

Fas

0.71

minist

0.66

RET

0.66

 Hawth

0.66

 Albion

0.66

 âĢİ

0.66

 Motorsport

0.64

Activations Density 0.028%

phrases related to viewpoints or perspectives

references to different perspectives or viewpoints

No Comments

No Known Activations