INDEX

Explanations

instances of the word "Jane" followed by a name or surname

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Jane" or variations thereof

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

ctic

-1.01

*/(

-0.97

iated

-0.81

idated

-0.75

Disclaimer

-0.73

orescent

-0.72

ulators

-0.71

natureconservancy

-0.70

è¦ļéĨĴ

-0.70

fecture

-0.66

POSITIVE LOGITS

Doe

1.40

 Jacobs

1.00

 Aust

0.99

Jane

0.91

 Seymour

0.91

 Mayer

0.90

 Jane

0.88

Roe

0.87

 Foster

0.83

etta

0.83

Activations Density 0.023%

instances of the word "Jane" followed by a name or surname

mentions of the name "Jane" or variations thereof

No Comments

No Known Activations