INDEX

Explanations

proper nouns related to specific individuals

oai_token-act-pair · gpt-3.5-turbo

names of individuals or entities, particularly those involved in legal or crime-related contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

vier

-0.93

itarian

-0.76

ological

-0.70

Magikarp

-0.68

 Vatican

-0.66

 Incarnation

-0.65

Catholic

-0.65

onymous

-0.65

 Galile

-0.64

ISM

-0.64

POSITIVE LOGITS

McF

1.22

arlane

1.18

adden

1.05

ORD

0.91

ees

0.87

erent

0.87

entanyl

0.86

ork

0.83

iculty

0.83

err

0.82

Activations Density 0.019%

proper nouns related to specific individuals

names of individuals or entities, particularly those involved in legal or crime-related contexts

No Comments

No Known Activations