INDEX

Explanations

action-related phrases related to setting up for a task or activity

oai_token-act-pair · gpt-3.5-turbo

expressions related to productivity and involvement in activities

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

-0.56

 convol

-0.55

pse

-0.54

 incorpor

-0.53

 Leilan

-0.53

chev

-0.51

Downloadha

-0.50

 anecd

-0.50

 hemor

-0.49

ãĤ©

-0.48

POSITIVE LOGITS

 RELEASE

0.60

 ASAP

0.48

 Xperia

0.48

IRE

0.45

 Halo

0.45

 Destiny

0.44

 FREE

0.43

 Posted

0.42

 Image

0.42

 Farming

0.42

Activations Density 1.424%

action-related phrases related to setting up for a task or activity

expressions related to productivity and involvement in activities

No Comments

No Known Activations