INDEX

Explanations

information about individuals' family backgrounds and origins

oai_token-act-pair · gpt-3.5-turbo

references to personal backgrounds or family origins

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

-0.81

soever

-0.73

Cooldown

-0.72

etheus

-0.69

cific

-0.69

 CrossRef

-0.69

*/(

-0.69

Recommend

-0.68

ultimate

-0.67

morrow

-0.67

POSITIVE LOGITS

 parents

0.96

 upbringing

0.96

 wealthy

0.94

 affluent

0.92

 impoverished

0.92

 arist

0.92

 immigrant

0.92

 poverty

0.91

 rural

0.90

 farming

0.90

Activations Density 0.276%

information about individuals' family backgrounds and origins

references to personal backgrounds or family origins

No Comments

No Known Activations