INDEX

Explanations

mentions of military forces and troop movements

oai_token-act-pair · gpt-3.5-turbo

references to military troop movements and deployments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Publishers

-0.84

coin

-0.83

Consumer

-0.81

 sinners

-0.78

writers

-0.76

writer

-0.72

Publisher

-0.72

 Dish

-0.71

 Yelp

-0.71

Bitcoin

-0.71

POSITIVE LOGITS

 deployments

1.52

 deployment

1.43

 troop

1.43

 deployed

1.40

 NATO

1.39

 rede

1.38

 garrison

1.33

 stationed

1.32

 deploy

1.30

 Nato

1.29

Activations Density 0.416%

mentions of military forces and troop movements

references to military troop movements and deployments

No Comments

No Known Activations