INDEX

Explanations

phrases related to war and conflict

oai_token-act-pair · gpt-3.5-turbo

terms related to wartime and postwar contexts

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

agne

-0.90

OGR

-0.90

ongs

-0.89

adr

-0.89

wagen

-0.88

omed

-0.85

uana

-0.84

afia

-0.81

ACP

-0.79

ertodd

-0.79

POSITIVE LOGITS

 contingency

1.05

 wartime

1.01

etheless

0.94

 staging

0.72

 blot

0.70

 caution

0.70

 fluorescent

0.68

 wardrobe

0.68

 nightmare

0.67

 Ministry

0.67

Activations Density 0.018%

phrases related to war and conflict

terms related to wartime and postwar contexts

No Comments

No Known Activations