INDEX

Explanations

proactive subscription requests or opt-ins

oai_token-act-pair · gpt-3.5-turbo

phrases related to receiving subscriptions and updates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Depth

-0.65

hatt

-0.64

pan

-0.63

 Bengal

-0.62

 deline

-0.60

sect

-0.60

 Tsukuyomi

-0.60

hole

-0.60

 mortg

-0.59

auld

-0.59

POSITIVE LOGITS

 notifications

0.85

 payments

0.81

 invitations

0.79

 notification

0.78

 gifts

0.77

ILY

0.76

imental

0.75

 periodic

0.75

olicited

0.73

soever

0.72

Activations Density 0.023%

proactive subscription requests or opt-ins

phrases related to receiving subscriptions and updates

No Comments

No Known Activations