INDEX

Explanations

phrases related to prisons or incarceration

oai_token-act-pair · gpt-3.5-turbo

mentions of prisons and incarceration

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

ï¸ı

-0.69

manship

-0.68

éĢ

-0.68

tz

-0.67

Sty

-0.66

Sean

-0.66

drops

-0.65

Ry

-0.64

Present

-0.63

Attributes

-0.63

POSITIVE LOGITS

 prison

3.59

 prisons

2.71

 jail

2.71

 Prison

2.67

prison

2.53

 correctional

2.12

 incarceration

2.10

 prisoners

2.06

 prisoner

2.06

 Correctional

2.02

Activations Density 0.014%

phrases related to prisons or incarceration

mentions of prisons and incarceration

No Comments

No Known Activations