INDEX

Explanations

the word "notify" or its variations

oai_token-act-pair · gpt-3.5-turbo

phrases related to the concept of notification or informing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

imb

-0.70

puff

-0.68

fan

-0.67

nic

-0.66

racted

-0.66

 alley

-0.64

cell

-0.62

 Huck

-0.61

cheon

-0.61

mun

-0.60

POSITIVE LOGITS

isance

0.99

 subscribers

0.92

 notify

0.85

 Subscribe

0.85

 notified

0.79

 Leilan

0.78

tale

0.76

 notifications

0.75

ĵĺ

0.75

ysis

0.74

Activations Density 0.022%

the word "notify" or its variations

phrases related to the concept of notification or informing

No Comments

No Known Activations