INDEX

Explanations

mentions or expressions of intense negative emotions such as hatred

oai_token-act-pair · gpt-3.5-turbo

expressions of hatred towards various groups or issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

UNCH

-0.79

debian

-0.79

ODE

-0.75

aver

-0.73

umm

-0.73

annis

-0.72

Interstitial

-0.71

ead

-0.69

arist

-0.68

AMA

-0.67

POSITIVE LOGITS

 hatred

1.08

 prejudice

1.01

 vengeance

0.87

 towards

0.84

fully

0.81

 hate

0.78

ãĥĨ

0.78

 intolerance

0.77

 prejudices

0.76

 animosity

0.76

Activations Density 0.018%

mentions or expressions of intense negative emotions such as hatred

expressions of hatred towards various groups or issues

No Comments

No Known Activations