INDEX

Explanations

references to the United States military and government

oai_token-act-pair · gpt-3.5-turbo

references to the United States and its military

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

ãĤ¦ãĤ¹

-0.93

ãĤ½

-0.81

phies

-0.79

ãĤ¡

-0.78

bably

-0.76

ãĥ¼ãĥ«

-0.74

 Emin

-0.73

sbm

-0.72

ĵĺ

-0.71

ãĤ¨ãĥ«

-0.70

POSITIVE LOGITS

 Department

1.10

 Geological

1.07

 government

1.05

 Embassy

1.05

 Postal

1.05

 Government

1.01

 embassy

0.96

 Bureau

0.93

 National

0.92

 judiciary

0.92

Activations Density 0.197%

references to the United States military and government

references to the United States and its military

No Comments

No Known Activations