INDEX

Explanations

terms related to addictions and addiction treatment

oai_token-act-pair · gpt-3.5-turbo

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

vard

-0.73

Kov

-0.72

erial

-0.69

dain

-0.68

arnaev

-0.67

thren

-0.67

asonic

-0.66

semble

-0.65

bourg

-0.65

osaurs

-0.65

POSITIVE LOGITS

 relapse

1.15

 addiction

1.10

 disorder

1.00

 Addiction

0.99

 disorders

0.98

 addicts

0.97

 addict

0.96

 rehab

0.95

 medication

0.88

 rehabilitation

0.87

Activations Density 0.048%

terms related to addictions and addiction treatment

No Comments

No Known Activations