INDEX

Explanations

mentions of specific threats, especially related to harassment and death

oai_token-act-pair · gpt-3.5-turbo

the term "threats," particularly in contexts related to harm or intimidation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

mys

-0.87

arist

-0.83

ICAN

-0.76

ilts

-0.76

Mech

-0.74

ria

-0.70

NAS

-0.69

Bed

-0.69

Band

-0.69

coat

-0.68

POSITIVE LOGITS

 threats

1.16

 challeng

1.03

 threat

0.95

 proble

0.91

 undermin

0.86

 threatening

0.84

 Threat

0.82

 behav

0.81

 posed

0.81

 threaten

0.80

Activations Density 0.009%

mentions of specific threats, especially related to harassment and death

the term "threats," particularly in contexts related to harm or intimidation

No Comments

No Known Activations