INDEX

Explanations

words related to confinement or imprisonment, especially focusing on the concept of being behind bars

oai_token-act-pair · gpt-3.5-turbo

references to incarceration or prison

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ctive

-0.92

lihood

-0.74

 GENERAL

-0.73

ALLY

-0.70

UAL

-0.70

IBLE

-0.68

 STATES

-0.67

ãĤī

-0.65

SOS

-0.64

ULAR

-0.63

POSITIVE LOGITS

poon

1.09

hop

1.00

hops

0.96

becue

0.95

 bars

0.94

hirt

0.93

celona

0.91

bell

0.91

mith

0.89

itone

0.86

Activations Density 0.011%

words related to confinement or imprisonment, especially focusing on the concept of being behind bars

references to incarceration or prison

No Comments

No Known Activations