INDEX

Explanations

references to the Latino community

oai_token-act-pair · gpt-3.5-turbo

references to Latino identity and community

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

tyard

-0.86

uden

-0.83

manship

-0.83

lopp

-0.78

artment

-0.77

lessly

-0.75

awaru

-0.74

éĹĺ

-0.73

akura

-0.73

ffff

-0.73

POSITIVE LOGITS

 Nadu

0.87

pes

0.83

 immigrant

0.76

 immigrants

0.75

 Favor

0.73

 Latinos

0.72

 Latino

0.71

 youths

0.70

 nationalist

0.68

naire

0.67

Activations Density 0.014%

references to the Latino community

references to Latino identity and community

No Comments

No Known Activations