INDEX

Explanations

phrases indicating prohibition or restriction

oai_token-act-pair · gpt-3.5-turbo

phrases related to prohibitions and restrictions

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

livion

-0.73

utics

-0.66

Nice

-0.66

POL

-0.65

tons

-0.65

 Anon

-0.65

didn

-0.65

orld

-0.64

rather

-0.64

leans

-0.64

POSITIVE LOGITS

 exceed

1.24

 interfere

1.01

 affect

0.92

 discriminate

0.91

ifies

0.87

 modify

0.86

 violate

0.85

 transfer

0.84

 adversely

0.84

 incur

0.84

Activations Density 0.198%

phrases indicating prohibition or restriction

phrases related to prohibitions and restrictions

No Comments

No Known Activations