INDEX

Explanations

keywords and names related to political figures and events

oai_token-act-pair · gpt-3.5-turbo

words related to defiance and opposition

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Mem

-0.76

Champ

-0.74

 Gloss

-0.74

mem

-0.70

NetMessage

-0.69

 Paste

-0.69

 Louis

-0.67

 Manit

-0.65

embed

-0.65

mbuds

-0.65

POSITIVE LOGITS

atta

1.90

 disob

1.68

 disobedience

1.64

 defy

1.53

 defiance

1.48

 defied

1.39

 defiant

1.34

 obey

1.25

obe

1.23

 risky

1.19

Activations Density 0.062%

keywords and names related to political figures and events

words related to defiance and opposition

No Comments

No Known Activations