INDEX

Explanations

phrases related to spy activities and spy software

oai_token-act-pair · gpt-3.5-turbo

descriptions of spy-related products and their features

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

 Slate

-0.85

"—

-0.83

 Ramsay

-0.81

Advertisement

-0.80

—"

-0.80

Blake

-0.79

 Burke

-0.78

Scott

-0.78

 Whedon

-0.78

Jackson

-0.77

POSITIVE LOGITS

 resear

1.09

Â´

1.08

 till

1.04

 english

1.02

 happ

0.97

 analyse

0.97

 physic

0.95

 alot

0.92

 learnt

0.89

!!

0.89

Activations Density 1.990%

phrases related to spy activities and spy software

descriptions of spy-related products and their features

No Comments

No Known Activations