INDEX

Explanations

words related to passwords and login credentials

oai_token-act-pair · gpt-3.5-turbo

instances and references to passwords in various contexts

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

 Fathers

-0.77

icles

-0.75

rums

-0.74

 guiActiveUnfocused

-0.73

 Spirits

-0.69

 Galile

-0.69

ICLE

-0.68

Church

-0.68

icle

-0.67

 Yards

-0.66

POSITIVE LOGITS

 password

1.23

 passwords

1.21

 authentication

1.05

 login

1.04

password

1.02

 hashes

1.01

 Password

0.98

 prompt

0.95

 credentials

0.93

 hash

0.92

Activations Density 0.044%

words related to passwords and login credentials

instances and references to passwords in various contexts

No Comments

No Known Activations