INDEX

Explanations

terms related to military or legal investigations

oai_token-act-pair · gpt-3.5-turbo

important updates and announcements related to various topics

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

emale

-0.55

 referen

-0.54

çīĪ

-0.46

ä»£

-0.45

ãĤ©

-0.45

ãĥ¼ãĥĨãĤ£

-0.44

achine

-0.42

ñ

-0.41

 Lange

-0.40

Jol

-0.40

POSITIVE LOGITS

NetMessage

0.46

 harsher

0.45

sw

0.44

 obstruction

0.44

rha

0.43

arent

0.42

 sooner

0.42

astical

0.42

adiq

0.42

vind

0.41

Activations Density 5.179%

terms related to military or legal investigations

important updates and announcements related to various topics

No Comments

No Known Activations