INDEX

Explanations

words related to military and conflict settings

oai_token-act-pair · gpt-3.5-turbo

references to battlegrounds or significant conflict zones

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

oras

-0.92

omaly

-0.86

abet

-0.84

essee

-0.84

oline

-0.83

ola

-0.83

quer

-0.83

oway

-0.83

por

-0.82

unal

-0.82

POSITIVE LOGITS

halla

0.82

wip

0.67

 shenan

0.67

0.66

 antics

0.66

 chemistry

0.64

 Advantage

0.62

 readiness

0.62

 bargaining

0.61

 Ready

0.60

Activations Density 0.304%

words related to military and conflict settings

references to battlegrounds or significant conflict zones

No Comments

No Known Activations