INDEX

Explanations

technical terms related to software updates and configurations

oai_token-act-pair · gpt-3.5-turbo

references to processes and updates in technology or software

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

enium

-0.67

 Dating

-0.60

 Tours

-0.58

ottenham

-0.57

 adolesc

-0.55

Atl

-0.53

 Evangel

-0.53

aspberry

-0.53

ç¥ŀ

-0.52

Ö

-0.52

POSITIVE LOGITS

 hazard

0.62

 payable

0.59

 incentiv

0.58

 ensured

0.58

 abound

0.56

anwhile

0.56

 stip

0.56

 comprom

0.55

 aided

0.54

 meanwhile

0.53

Activations Density 2.626%

technical terms related to software updates and configurations

references to processes and updates in technology or software

No Comments

No Known Activations