INDEX

Explanations

phrases related to consequences or conditions

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate conditions or requirements associated with various subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Discuss

-0.78

chan

-0.76

emis

-0.76

zan

-0.75

atism

-0.73

antis

-0.73

OWER

-0.71

nesota

-0.71

unker

-0.70

rn

-0.69

POSITIVE LOGITS

 caveats

1.20

 baggage

1.08

 disclaim

1.08

 caveat

0.95

 disclaimer

0.93

 warranty

0.90

 strings

0.89

 stip

0.88

 handy

0.87

 accompanying

0.87

Activations Density 0.128%

phrases related to consequences or conditions

phrases that indicate conditions or requirements associated with various subjects

No Comments

No Known Activations