INDEX

Explanations

references to dining locations

oai_token-act-pair · gpt-3.5-turbo

references to dining experiences and related establishments

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

 Archangel

-0.81

Mub

-0.70

 annexed

-0.68

 Aerospace

-0.68

 Belarus

-0.67

zyk

-0.66

 Mehran

-0.66

Downloadha

-0.64

ality

-0.63

 Pryor

-0.62

POSITIVE LOGITS

 halls

0.97

 dining

0.93

cloth

0.88

din

0.88

eteria

0.87

 lounge

0.83

iasco

0.82

 tastes

0.80

 etiquette

0.79

 establishments

0.79

Activations Density 0.022%

references to dining locations

references to dining experiences and related establishments

No Comments

No Known Activations