INDEX

Explanations

proper nouns, specifically names of locations, organizations, and job titles

oai_token-act-pair · gpt-3.5-turbo

phrases indicating ownership or association

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

 accordingly

-0.96

 thereafter

-0.88

 thereof

-0.86

 thereto

-0.82

/,

-0.81

*.

-0.80

 whereas

-0.79

ãĢĤ

-0.78

+.

-0.77

 therein

-0.76

POSITIVE LOGITS

 Massachusetts

0.79

 embattled

0.78

 Melbourne

0.77

 Pennsylvania

0.77

 Alabama

0.76

 Georgia

0.75

 Malaysian

0.74

 Oklahoma

0.73

 Palestinian

0.73

 Florida

0.73

Activations Density 0.403%

proper nouns, specifically names of locations, organizations, and job titles

phrases indicating ownership or association

No Comments

No Known Activations