INDEX

Explanations

mentions of government ministries or departments

oai_token-act-pair · gpt-3.5-turbo

terms related to governmental and administrative entities

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

 Vermont

-0.88

 Courtney

-0.82

 Delaware

-0.81

 Bowman

-0.79

 Oregon

-0.77

 Amtrak

-0.76

 Drake

-0.76

 Wisconsin

-0.75

 Nebraska

-0.75

 Mormon

-0.75

POSITIVE LOGITS

ÄŁ

1.21

 Directorate

1.04

ariat

0.97

 Ankara

0.95

fecture

0.95

oglu

0.93

ÅŁ

0.92

 palace

0.87

 delegation

0.86

 Ministry

0.86

Activations Density 0.359%

mentions of government ministries or departments

terms related to governmental and administrative entities

No Comments

No Known Activations