INDEX

Explanations

mentions of governmental or official roles or activities

oai_token-act-pair · gpt-3.5-turbo

references to governmental and military structures or roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

rame

-1.02

bnb

-0.98

Downloadha

-0.83

ivo

-0.82

HD

-0.81

ï¸ı

-0.79

Update

-0.79

videos

-0.79

isode

-0.79

(@

-0.78

POSITIVE LOGITS

 Soviets

1.44

 USSR

1.33

 Soviet

1.29

 Ottoman

1.28

 Confederacy

1.22

 Bolshe

1.20

 Germans

1.20

 colonists

1.20

 Nazis

1.19

 postwar

1.16

Activations Density 0.524%

mentions of governmental or official roles or activities

references to governmental and military structures or roles

No Comments

No Known Activations