INDEX

Explanations

mentions of celebrities and their activities or appearances

oai_token-act-pair · gpt-3.5-turbo

mentions of celebrity performances and public appearances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Recommend

-0.85

Information

-0.84

 Restrict

-0.76

database

-0.74

soType

-0.74

itutional

-0.69

information

-0.69

 reliance

-0.68

 vacancies

-0.68

estate

-0.67

POSITIVE LOGITS

 strut

1.41

 onstage

1.28

 grinning

1.24

 gigg

1.22

 hilar

1.21

 danced

1.21

 backstage

1.19

 effortlessly

1.16

 selfie

1.16

 cheers

1.16

Activations Density 0.718%

mentions of celebrities and their activities or appearances

mentions of celebrity performances and public appearances

No Comments

No Known Activations