INDEX

Explanations

sentences expressing emotional intensity or urgency

oai_token-act-pair · gpt-3.5-turbo

special characters or symbols used in text

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 redist

-0.71

ANC

-0.69

 monet

-0.68

 demographics

-0.68

 certification

-0.67

 Communities

-0.67

 Bloomberg

-0.67

 Networks

-0.65

 distribut

-0.65

 GMOs

-0.64

POSITIVE LOGITS

¹

1.27

Ń

1.22

»

1.19

£

1.18

¡

1.16

½

1.14

º

1.14

ª

1.14

¦

1.09

Ķ

1.08

Activations Density 0.237%

sentences expressing emotional intensity or urgency

special characters or symbols used in text

No Comments

No Known Activations