INDEX

Explanations

references to costumes

oai_token-act-pair · gpt-3.5-turbo

references to costumes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

upon

-0.82

-0.73

venants

-0.73

tical

-0.72

raltar

-0.67

orio

-0.66

avanaugh

-0.65

utherford

-0.65

Reviewed

-0.63

Hes

-0.62

POSITIVE LOGITS

 costumes

1.09

 costume

1.07

 attire

0.86

 Costume

0.86

 jewelry

0.84

 wardrobe

0.79

 apparel

0.77

ÃįÃį

0.77

 gown

0.76

 wearer

0.75

Activations Density 0.014%

references to costumes

references to costumes

No Comments

No Known Activations