INDEX

Explanations

statements indicating a morally or practically "right" action or decision

oai_token-act-pair · gpt-3.5-turbo

phrases indicating correctness or justification in decision-making

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

gyn

-0.85

stories

-0.85

 Lists

-0.80

olics

-0.79

ographics

-0.77

 Races

-0.76

fam

-0.75

bows

-0.73

images

-0.72

»Ĵ

-0.72

POSITIVE LOGITS

 move

1.46

 decision

1.44

 approach

1.34

 step

1.34

 course

1.29

 choice

1.28

 tactic

1.25

 option

1.24

 strategy

1.23

 stance

1.20

Activations Density 0.226%

statements indicating a morally or practically "right" action or decision

phrases indicating correctness or justification in decision-making

No Comments

No Known Activations