INDEX

Explanations

phrases related to instructional or how-to content

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 implying

-0.69

uploads

-0.66

 interven

-0.66

ighed

-0.66

 reiter

-0.66

 Defendants

-0.65

 repud

-0.65

 entirety

-0.65

alion

-0.64

 Plaint

-0.64

POSITIVE LOGITS

 Yourself

1.07

 yourself

1.00

 Your

0.94

 safely

0.94

 your

0.91

 securely

0.82

 YOUR

0.77

 Proper

0.77

imum

0.76

successful

0.75

Activations Density 0.520%

phrases related to instructional or how-to content

No Comments

No Known Activations