INDEX

Explanations

phrases related to political/military deterrence

oai_token-act-pair · gpt-3.5-turbo

concepts related to deterrence, particularly in the context of nuclear strategy

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

 Printed

-0.89

apor

-0.75

gres

-0.74

ao

-0.72

eds

-0.71

annis

-0.71

 Created

-0.71

andre

-0.71

paces

-0.70

esta

-0.70

POSITIVE LOGITS

 deterrent

1.75

 deterrence

1.46

 enthusi

0.93

 juven

0.85

 scourge

0.80

 outwe

0.78

 conclud

0.77

 deter

0.74

 dissu

0.73

 reinforcement

0.73

Activations Density 0.008%

phrases related to political/military deterrence

concepts related to deterrence, particularly in the context of nuclear strategy

No Comments

No Known Activations