INDEX

Explanations

references to suicidal actions or thoughts

oai_token-act-pair · gpt-3.5-turbo

references to suicidal individuals and suicidal behavior

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 apparel

-0.75

Rew

-0.71

 Corn

-0.70

iture

-0.69

 Renew

-0.69

orns

-0.68

itures

-0.66

 Foss

-0.65

Vel

-0.65

Prov

-0.65

POSITIVE LOGITS

 suicidal

2.65

 autistic

2.11

 delusional

1.96

 bipolar

1.92

 psychotic

1.89

 paranoid

1.84

 narcissistic

1.72

 schizophren

1.62

 sociop

1.59

icidal

1.54

Activations Density 0.053%

references to suicidal actions or thoughts

references to suicidal individuals and suicidal behavior

No Comments

No Known Activations