INDEX

Explanations

words related to social media sharing and interaction

oai_token-act-pair · gpt-3.5-turbo

references to social media engagement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ãĤ¡

-0.74

istically

-0.73

quartered

-0.71

pler

-0.64

tis

-0.63

mond

-0.62

 sovere

-0.62

 situ

-0.62

stress

-0.62

 accur

-0.61

POSITIVE LOGITS

 SHARES

1.16

 Facebook

1.12

 VIDEOS

1.03

 Share

1.03

 Email

0.96

 facebook

0.96

Pinterest

0.95

MSN

0.95

 Shares

0.92

facebook

0.89

Activations Density 0.019%

words related to social media sharing and interaction

references to social media engagement

No Comments

No Known Activations