INDEX

Explanations

references to video game titles, particularly those related to action and warfare

oai_token-act-pair · gpt-3.5-turbo

references to popular video game franchises

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

 clot

-0.72

upper

-0.72

mean

-0.69

agogue

-0.69

 filib

-0.69

conserv

-0.68

aff

-0.66

rss

-0.66

indu

-0.66

rist

-0.66

POSITIVE LOGITS

 Royale

1.07

 Kinect

1.02

 Arcade

1.00

 Ubisoft

0.92

Xbox

0.92

 gameplay

0.91

 Multiplayer

0.90

 Trilogy

0.88

 XCOM

0.88

 multiplayer

0.87

Activations Density 0.129%

references to video game titles, particularly those related to action and warfare

references to popular video game franchises

No Comments

No Known Activations