INDEX

Explanations

phrases indicating a call to action or directive to do something

oai_token-act-pair · gpt-3.5-turbo

imperative commands

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

 advertised

-0.69

eers

-0.68

Cong

-0.65

 david

-0.65

loo

-0.64

iege

-0.63

nesses

-0.63

 agre

-0.62

tem

-0.61

 brill

-0.60

POSITIVE LOGITS

aways

1.23

 advantage

0.98

overs

0.97

 aback

0.93

 heed

0.84

away

0.79

OVER

0.78

YR

0.78

autions

0.77

ume

0.77

Activations Density 0.105%

phrases indicating a call to action or directive to do something

imperative commands

No Comments

No Known Activations