INDEX

Explanations

individuals in positions of authority or responsibility

oai_token-act-pair · gpt-3.5-turbo

words associated with different roles or professions

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

 Enemies

-0.73

 Presidents

-0.71

 Monsters

-0.68

 Pieces

-0.64

DragonMagazine

-0.64

 Governors

-0.62

>[

-0.62

çķ

-0.62

 Silence

-0.61

æĪ¦

-0.61

POSITIVE LOGITS

 named

1.33

 testified

1.25

 remarked

1.20

 told

1.17

 surn

1.16

 wrote

1.15

 commented

1.15

 overheard

1.14

 complained

1.11

 nicknamed

1.07

Activations Density 0.164%

individuals in positions of authority or responsibility

words associated with different roles or professions

No Comments

No Known Activations