INDEX

Explanations

words indicating likelihood or unlikelihood of events happening

oai_token-act-pair · gpt-3.5-turbo

phrases expressing likelihood or unlikelihood

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ussion

-0.73

avorite

-0.73

aband

-0.73

utch

-0.66

aving

-0.65

jab

-0.64

Ã§a

-0.63

afia

-0.63

Mau

-0.63

athed

-0.62

POSITIVE LOGITS

 probable

0.68

 underest

0.67

 underestimate

0.66

 exaggeration

0.65

 implied

0.65

 circumst

0.64

 evolution

0.64

 conceivable

0.63

imaru

0.63

arg

0.63

Activations Density 0.089%

words indicating likelihood or unlikelihood of events happening

phrases expressing likelihood or unlikelihood

No Comments

No Known Activations