INDEX

Explanations

references to official visits and trips

oai_token-act-pair · gpt-3.5-turbo

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

Pru

-0.58

corn

-0.56

Ü

-0.56

 pulp

-0.56

Apps

-0.56

fry

-0.55

 breaching

-0.54

acebook

-0.54

Jou

-0.53

theless

-0.53

POSITIVE LOGITS

 abroad

1.02

acion

0.95

ations

0.87

to

0.81

cation

0.79

ees

0.78

 visits

0.76

orial

0.76

PsyNetMessage

0.74

è¦ļéĨĴ

0.72

Activations Density 0.039%

references to official visits and trips

No Comments

No Known Activations