INDEX

Explanations

words related to swearing or using profanity

oai_token-act-pair · gpt-3.5-turbo

instances of the word "swear" and its variations

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

issues

-0.83

ItemTracker

-0.77

Cent

-0.70

Center

-0.69

issue

-0.69

Zen

-0.69

 guiActiveUnfocused

-0.69

iga

-0.68

EMS

-0.67

neau

-0.67

POSITIVE LOGITS

 swear

1.41

 swearing

1.25

 swore

1.14

 oath

1.09

 sworn

1.01

thouse

0.86

 warr

0.83

 confir

0.81

 agre

0.80

 ingred

0.79

Activations Density 0.006%

words related to swearing or using profanity

instances of the word "swear" and its variations

No Comments

No Known Activations