INDEX

Explanations

terms related to specific categories or topics such as healthcare, politics, and economy

oai_token-act-pair · gpt-3.5-turbo

specific nouns and terms related to various categories and contexts

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

ĸļ

-0.96

 Shutterstock

-0.73

 Archdemon

-0.67

 Townsend

-0.64

 Rasmussen

-0.64

terday

-0.64

hement

-0.63

 Carnegie

-0.63

TNT

-0.62

 Debor

-0.62

POSITIVE LOGITS

 endorsements

0.79

 categories

0.79

related

0.76

 types

0.74

 acron

0.73

 outings

0.72

 variants

0.72

 titles

0.72

 denominations

0.70

isations

0.70

Activations Density 0.633%

terms related to specific categories or topics such as healthcare, politics, and economy

specific nouns and terms related to various categories and contexts

No Comments

No Known Activations