INDEX

Explanations

medical procedures

oai_token-act-pair · gpt-3.5-turbo

phrases indicating actions taken by someone or something

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

SpaceEngineers

-0.85

 Hail

-0.81

inventoryQuantity

-0.80

verting

-0.74

 Kurd

-0.74

Ply

-0.64

 Speech

-0.64

ctuary

-0.64

 Patriot

-0.63

itating

-0.62

POSITIVE LOGITS

alore

0.76

BRE

0.71

rium

0.70

 electroly

0.69

Cod

0.68

eria

0.67

 reader

0.67

issance

0.66

LER

0.65

stro

0.64

Activations Density 0.000%

medical procedures

phrases indicating actions taken by someone or something

No Comments

No Known Activations