INDEX

Explanations

references to isolation or solitary conditions

oai_token-act-pair · gpt-3.5-turbo

references to solitary confinement and related concepts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oppers

-0.75

rote

-0.69

ressor

-0.67

nant

-0.67

ound

-0.66

enegger

-0.66

Ã©n

-0.65

Mass

-0.64

rosc

-0.63

orthy

-0.63

POSITIVE LOGITS

 confinement

1.57

 solitary

1.44

theless

0.77

spection

0.75

 charact

0.75

 lone

0.73

 prosecutions

0.71

uously

0.68

 sqor

0.68

 pens

0.68

Activations Density 0.008%

references to isolation or solitary conditions

references to solitary confinement and related concepts

No Comments

No Known Activations