INDEX

Explanations

phrases related to completing tasks or projects

oai_token-act-pair · gpt-3.5-turbo

phrases related to strong emotional expressions or experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Vaugh

-0.67

enegger

-0.63

 helicop

-0.60

 Moroc

-0.56

oppable

-0.56

bledon

-0.56

 disadvant

-0.55

 Frie

-0.52

 incorpor

-0.50

 Niet

-0.50

POSITIVE LOGITS

\":

0.63

hs

0.47

photos

0.45

 splash

0.44

lder

0.44

 Halo

0.43

 teasing

0.42

':

0.41

Â¶

0.41

rots

0.41

Activations Density 2.165%

phrases related to completing tasks or projects

phrases related to strong emotional expressions or experiences

No Comments

No Known Activations