INDEX

Explanations

phrases related to different perspectives or viewpoints

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 @ 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wen

-0.77

wyn

-0.68

pload

-0.68

sounding

-0.66

knit

-0.65

KEN

-0.64

zzy

-0.63

ESSION

-0.63

ayn

-0.62

conn

-0.62

POSITIVE LOGITS

aries

0.77

ports

0.75

atics

0.75

finder

0.75

 viewpoint

0.72

ership

0.72

 perspectives

0.70

 perspective

0.70

 mirrors

0.69

 bulbs

0.69

Activations Density 0.015%

phrases related to different perspectives or viewpoints

references to different perspectives or viewpoints

No Comments

No Known Activations