INDEX

Explanations

mentions of weapons

oai_token-act-pair · gpt-3.5-turbo

references to weapons and armaments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

noon

-0.76

 Strawberry

-0.75

vent

-0.73

 Woodward

-0.73

gres

-0.72

lus

-0.70

orse

-0.70

Adams

-0.68

athan

-0.68

âĸ¬

-0.66

POSITIVE LOGITS

 weapons

0.96

 smugglers

0.92

mith

0.91

 caches

0.90

 arsenal

0.87

powder

0.87

 shipments

0.87

 smuggling

0.86

manship

0.84

ensing

0.83

Activations Density 0.022%

mentions of weapons

references to weapons and armaments

No Comments

No Known Activations