INDEX

Explanations

mentions of military or combat-related terms and activities

oai_token-act-pair · gpt-3.5-turbo

mentions of specific groups or teams, particularly those associated with violence or conflict

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 tongue

-0.71

ocaust

-0.69

gement

-0.67

bir

-0.66

 cherished

-0.66

issance

-0.65

 fetish

-0.64

po

-0.64

ously

-0.64

JUST

-0.63

POSITIVE LOGITS

 Squad

1.31

rons

1.12

arnaev

0.83

atile

0.81

 Squadron

0.81

 Orchestra

0.79

IDs

0.78

 Participation

0.78

Corp

0.77

Squ

0.75

Activations Density 0.006%

mentions of military or combat-related terms and activities

mentions of specific groups or teams, particularly those associated with violence or conflict

No Comments

No Known Activations