INDEX

Explanations

phrases related to internet user behavior and online community management

oai_token-act-pair · gpt-3.5-turbo

references to spam and platform moderation policies

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

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.92

 Clarke

-0.80

 Jennings

-0.77

 Bundy

-0.76

 Desmond

-0.76

 ensemble

-0.74

 Quart

-0.73

 Calder

-0.71

 Roland

-0.69

 Bauer

-0.69

POSITIVE LOGITS

 spam

1.85

 scams

1.34

 abusive

1.25

 scam

1.25

 bots

1.24

 harassing

1.24

 slander

1.23

 malicious

1.20

 imperson

1.17

 annoying

1.17

Activations Density 0.464%

phrases related to internet user behavior and online community management

references to spam and platform moderation policies

No Comments

No Known Activations