INDEX

Explanations

URLs with '.uk' domain names

oai_token-act-pair · gpt-3.5-turbo

occurrences of the domain "co.uk" and related web addresses

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

milo

-0.75

values

-0.64

fixes

-0.63

hift

-0.63

idate

-0.62

busters

-0.62

coded

-0.62

tailed

-0.61

Ö¼

-0.59

goers

-0.59

POSITIVE LOGITS

/?

0.81

ithing

0.77

ibaba

0.75

arthed

0.72

MpServer

0.72

gypt

0.72

 Lumpur

0.70

miah

0.68

 âĢº

0.68

anse

0.67

Activations Density 0.022%

URLs with '.uk' domain names

occurrences of the domain "co.uk" and related web addresses

No Comments

No Known Activations