INDEX

Explanations

mentions of coffee-related words

oai_token-act-pair · gpt-3.5-turbo

references to coffee and coffee-related establishments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ROR

-0.86

oppable

-0.72

 Caldwell

-0.71

yss

-0.70

orse

-0.68

Filename

-0.65

avid

-0.63

alez

-0.61

etric

-0.61

ukong

-0.61

POSITIVE LOGITS

 beans

1.21

 Beans

1.05

 bean

1.00

 brewed

0.99

 drinkers

0.93

 cups

0.88

cake

0.87

weed

0.85

cakes

0.84

 drip

0.83

Activations Density 0.019%

mentions of coffee-related words

references to coffee and coffee-related establishments

No Comments

No Known Activations