INDEX

Explanations

words related to armed conflict and military actions

oai_token-act-pair · gpt-3.5-turbo

references to military actions and government responses

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Lovecraft

-0.93

Disney

-0.78

Seattle

-0.78

inburgh

-0.73

erella

-0.73

 Yelp

-0.73

uckland

-0.72

nesia

-0.70

phia

-0.70

 Cosponsors

-0.69

POSITIVE LOGITS

 mortar

1.39

 artillery

1.29

 snipers

1.24

 grenades

1.23

 sniper

1.22

 grenade

1.20

 gunfire

1.20

 airstrike

1.20

 ammunition

1.16

 armoured

1.16

Activations Density 0.324%

words related to armed conflict and military actions

references to military actions and government responses

No Comments

No Known Activations