INDEX

Explanations

references to users of a service or product

oai_token-act-pair · gpt-3.5-turbo

references to various subscriber counts and related metrics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

":"/

-0.89

kered

-0.67

Park

-0.66

«

-0.63

 bulldo

-0.63

 Allen

-0.63

park

-0.62

olith

-0.62

aldehyde

-0.62

 Bush

-0.62

POSITIVE LOGITS

 subscribers

3.82

 subscriber

3.26

 subscriptions

2.47

 subscription

1.99

 subscribed

1.75

 listeners

1.70

 subscribing

1.70

 subscribe

1.54

 subscrib

1.52

 readers

1.51

Activations Density 0.010%

references to users of a service or product

references to various subscriber counts and related metrics

No Comments

No Known Activations