INDEX

Explanations

actions related to improvement or change

oai_token-act-pair · gpt-3.5-turbo

references to progress and improvement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

don

-0.36

mun

-0.34

ãĥ³ãĤ¸

-0.34

 Cunning

-0.33

isl

-0.33

ãĥ¼ãĥĨãĤ£

-0.32

©¶æ

-0.32

equ

-0.31

 sole

-0.31

elve

-0.30

POSITIVE LOGITS

iatus

0.49

natureconservancy

0.48

NetMessage

0.46

APD

0.44

atform

0.42

terness

0.41

ĵĺ

0.39

Enlarge

0.39

earable

0.39

arcer

0.39

Activations Density 3.816%

actions related to improvement or change

references to progress and improvement

No Comments

No Known Activations