INDEX

Explanations

proper nouns representing specific entities, most likely organizations or events

oai_token-act-pair · gpt-3.5-turbo

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

witch

-0.79

acid

-0.77

abuse

-0.75

ghazi

-0.74

âĹ¼

-0.73

Invalid

-0.72

iably

-0.71

 anyway

-0.70

ratom

-0.70

POSITIVE LOGITS

 goal

1.32

aim

1.23

 announcement

1.09

 centerpiece

1.07

 inaugural

1.03

 project

1.02

 objective

1.01

 initiative

1.01

 purpose

1.00

 trio

1.00

Activations Density 0.280%

proper nouns representing specific entities, most likely organizations or events

No Comments

No Known Activations