INDEX

Explanations

mentions of disabilities or individuals with disabilities

oai_token-act-pair · gpt-3.5-turbo

mentions of disability and related terms

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

akeru

-0.88

soType

-0.84

uin

-0.84

alg

-0.80

furt

-0.78

ablishment

-0.75

é¾įå¥ĳå£«

-0.73

soDeliveryDate

-0.72

NG

-0.72

cia

-0.69

POSITIVE LOGITS

 disabled

0.86

locked

0.78

 Disabled

0.75

 disability

0.73

 aback

0.70

 limbs

0.69

river

0.69

bay

0.69

 thyroid

0.67

 Nadu

0.67

Activations Density 0.014%

mentions of disabilities or individuals with disabilities

mentions of disability and related terms

No Comments

No Known Activations