INDEX

Explanations

words related to negative social behavior or mistreatment of individuals

oai_token-act-pair · gpt-3.5-turbo

mentions of harassment and related behaviors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

éĹĺ

-0.99

icts

-0.78

inet

-0.78

zyme

-0.75

ACTED

-0.73

chart

-0.72

arch

-0.71

lined

-0.71

essential

-0.70

shows

-0.70

POSITIVE LOGITS

 harass

0.91

 harassment

0.90

 harassing

0.87

 harassed

0.78

assment

0.76

 stalking

0.75

 accus

0.75

 tactics

0.73

ãĥĨ

0.71

ingly

0.67

Activations Density 0.026%

words related to negative social behavior or mistreatment of individuals

mentions of harassment and related behaviors

No Comments

No Known Activations