INDEX

Explanations

instances of words related to accomplishments or positive outcomes

oai_token-act-pair · gpt-3.5-turbo

references to various types of achievements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ca

-0.68

 Photoshop

-0.66

ondo

-0.66

 Printed

-0.65

olls

-0.64

cise

-0.62

©¶æ

-0.62

 wildfire

-0.60

pher

-0.59

 depos

-0.59

POSITIVE LOGITS

 achievement

1.54

 accomplishment

1.36

 attainment

1.24

 achie

1.23

 achievements

1.18

 Achievement

1.17

chieve

1.02

 Achievements

0.93

 feats

0.89

 accol

0.89

Activations Density 0.008%

instances of words related to accomplishments or positive outcomes

references to various types of achievements

No Comments

No Known Activations