INDEX

Explanations

words related to the lack of connection, intention, relationship, or indication

oai_token-act-pair · gpt-3.5-turbo

terms associated with relationships and intentions

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

eg

-0.73

ometimes

-0.59

estern

-0.58

initely

-0.58

sung

-0.58

rongh

-0.57

 Rounds

-0.57

everal

-0.56

igs

-0.55

Sig

-0.55

POSITIVE LOGITS

 whatsoever

1.96

nor

1.14

 anymore

1.00

 except

0.89

nor

0.87

galitarian

0.77

 bothered

0.75

ivable

0.73

¬¼

0.72

ÑĢ

0.72

Activations Density 0.195%

words related to the lack of connection, intention, relationship, or indication

terms associated with relationships and intentions

No Comments

No Known Activations