INDEX

Explanations

references to books or written works

oai_token-act-pair · gpt-3.5-turbo

references to books and their authors

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

fax

-0.84

ime

-0.78

outube

-0.71

vd

-0.71

udeb

-0.68

ct

-0.65

aples

-0.65

hesda

-0.65

DERR

-0.65

 webcam

-0.64

POSITIVE LOGITS

 Capitalism

1.24

 Principles

1.23

 Understanding

1.21

 Thinking

1.21

 Reasons

1.20

 Shattered

1.14

 Killing

1.12

 Saving

1.12

 Bringing

1.10

 Secrets

1.09

Activations Density 0.157%

references to books or written works

references to books and their authors

No Comments

No Known Activations