INDEX

Explanations

mentions of anti-something, often in a negative context

oai_token-act-pair · gpt-3.5-turbo

terms related to anti-establishment and anti-social movements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ODY

-0.76

ãĤ¼ãĤ¦ãĤ¹

-0.74

 Pelicans

-0.72

 Halls

-0.72

shire

-0.71

ĸļ

-0.70

 Curve

-0.70

 Sins

-0.69

hower

-0.68

 Remastered

-0.67

POSITIVE LOGITS

abortion

1.09

war

1.05

establishment

1.02

overty

0.96

anti

0.89

hero

0.88

actively

0.87

strate

0.83

union

0.81

digy

0.79

Activations Density 0.018%

mentions of anti-something, often in a negative context

terms related to anti-establishment and anti-social movements

No Comments

No Known Activations