INDEX

Explanations

references to actors or actresses

oai_token-act-pair · gpt-3.5-turbo

references to actors and actresses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

manac

-0.74

cling

-0.70

rotein

-0.68

heet

-0.67

otide

-0.66

tical

-0.64

udge

-0.63

aq

-0.63

gn

-0.62

jong

-0.61

POSITIVE LOGITS

 actors

3.84

 actor

2.75

 actresses

2.30

Actor

1.98

 performers

1.95

 Actor

1.93

 actress

1.86

 filmmakers

1.77

 singers

1.66

 comedians

1.59

Activations Density 0.022%

references to actors or actresses

references to actors and actresses

No Comments

No Known Activations