INDEX

Explanations

phrases related to smoking and habits

oai_token-act-pair · gpt-3.5-turbo

references to smoking, health risks, and addiction-related behavior

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

Assembly

-0.80

 Barron

-0.78

ilater

-0.77

 Gould

-0.76

 CASE

-0.73

 FEMA

-0.71

arte

-0.71

Crew

-0.70

 Asylum

-0.70

 Cosponsors

-0.69

POSITIVE LOGITS

 smoking

1.35

 binge

1.31

 cigarettes

1.24

 smoker

1.24

 smokers

1.24

 cigarette

1.17

 habitual

1.16

 abstinence

1.16

 smoked

1.16

 abst

1.16

Activations Density 0.325%

phrases related to smoking and habits

references to smoking, health risks, and addiction-related behavior

No Comments

No Known Activations