INDEX

Explanations

phrases related to medical disorders

oai_token-act-pair · gpt-3.5-turbo

references to mental health disorders

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Pil

-0.70

aiden

-0.69

da

-0.66

riel

-0.64

ammy

-0.64

sson

-0.63

elines

-0.63

ven

-0.63

vel

-0.63

anga

-0.62

POSITIVE LOGITS

 disorder

1.21

 disorders

1.09

 Disorders

0.96

 Disorder

0.91

 psychiat

0.89

 worsen

0.86

 epile

0.82

 psychosis

0.81

 addicts

0.75

 wors

0.74

Activations Density 0.009%

phrases related to medical disorders

references to mental health disorders

No Comments

No Known Activations