INDEX

Explanations

dates and names written in a specific format

oai_token-act-pair · gpt-3.5-turbo

numerical data or references to dates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 disciplines

-0.84

 larvae

-0.84

 cons

-0.82

 governed

-0.77

 undet

-0.77

 incub

-0.76

 regener

-0.76

 covenant

-0.76

 puberty

-0.74

 commissions

-0.74

POSITIVE LOGITS

Loading

1.35

Advertisements

1.32

Translation

1.23

Cong

1.22

Looks

1.21

Calling

1.20

Though

1.20

Warning

1.20

RIP

1.20

Dear

1.20

Activations Density 0.139%

dates and names written in a specific format

numerical data or references to dates

No Comments

No Known Activations