INDEX

Explanations

phrases related to user authentication or logging into accounts

oai_token-act-pair · gpt-3.5-turbo

phrases related to logging in to accounts or services

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 tract

-0.70

bou

-0.70

 Galile

-0.65

 Holt

-0.64

 economical

-0.64

 Archbishop

-0.64

 diminishing

-0.63

 hect

-0.62

 Yards

-0.62

 Produ

-0.61

POSITIVE LOGITS

 credentials

1.34

 login

1.26

 Username

1.21

login

1.15

 Login

1.13

 password

1.09

password

1.09

 username

1.05

 Password

1.01

ername

1.01

Activations Density 0.126%

phrases related to user authentication or logging into accounts

phrases related to logging in to accounts or services

No Comments

No Known Activations