INDEX

Explanations

invalid email addresses

oai_token-act-pair · gpt-3.5-turbo

references to invalid or incorrect email addresses

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

senal

-0.71

ccording

-0.70

oggle

-0.63

 Akin

-0.59

bler

-0.59

hung

-0.58

against

-0.57

BIL

-0.56

 Guth

-0.55

ppard

-0.55

POSITIVE LOGITS

 captcha

1.05

 email

1.04

 username

0.88

$$$$

0.77

Email

0.74

 Username

0.73

icio

0.70

 emails

0.69

 password

0.67

 login

0.66

Activations Density 0.017%

invalid email addresses

references to invalid or incorrect email addresses

No Comments

No Known Activations