INDEX

Explanations

words related to violations of laws and regulations

oai_token-act-pair · gpt-3.5-turbo

references to various types of laws and legal violations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ggles

-0.64

otal

-0.63

chie

-0.61

optim

-0.60

pse

-0.59

apest

-0.58

ortun

-0.56

bitious

-0.56

rowth

-0.55

rir

-0.55

POSITIVE LOGITS

 prohibiting

1.17

 applicable

1.07

 governing

1.00

 prohibitions

1.00

 prohibits

0.95

 rules

0.94

 prohibition

0.91

 norms

0.91

 barring

0.91

 forb

0.91

Activations Density 0.113%

words related to violations of laws and regulations

references to various types of laws and legal violations

No Comments

No Known Activations