INDEX

Explanations

words related to proper nouns or names of individuals and places

oai_token-act-pair · gpt-3.5-turbo

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Masquerade

-0.87

 Sabb

-0.87

Tec

-0.85

Mub

-0.84

 Stab

-0.78

 Curiosity

-0.77

Pes

-0.75

 Schwe

-0.75

 Schultz

-0.72

 Chong

-0.72

POSITIVE LOGITS

or

1.45

OR

1.40

oris

1.30

ors

1.27

orians

1.22

oros

1.22

orian

1.19

ori

1.18

oring

1.18

oria

1.17

Activations Density 0.195%

words related to proper nouns or names of individuals and places

No Comments

No Known Activations