INDEX

Explanations

instances of people being insulted or making insulting comments

oai_token-act-pair · gpt-3.5-turbo

phrases related to social media interactions and conflicts

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

 Zucker

-0.96

Zub

-0.88

 dives

-0.85

suc

-0.81

SQ

-0.78

bles

-0.78

-0.76

 Perez

-0.73

 kittens

-0.73

 dive

-0.72

POSITIVE LOGITS

Arm

2.19

Arm

2.12

arm

2.08

ARM

2.03

ARM

1.83

arm

1.75

arms

1.45

 Armory

1.38

 Armstrong

1.38

 Arms

1.36

Activations Density 0.233%

instances of people being insulted or making insulting comments

phrases related to social media interactions and conflicts

No Comments

No Known Activations