INDEX

Explanations

phrases related to political, historical, and governmental entities

oai_token-act-pair · gpt-3.5-turbo

references to espionage and political intrigue

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

-0.62

 cheapest

-0.57

:(

-0.56

LOW

-0.56

 Awakens

-0.56

 Mania

-0.56

ISO

-0.54

upd

-0.54

 ðŁĺ

-0.54

 ðŁĻĤ

-0.53

POSITIVE LOGITS

 boycot

0.76

 universities

0.71

 scandals

0.67

 academia

0.64

 presided

0.64

 denouncing

0.62

 denounced

0.62

 whom

0.62

SPONSORED

0.62

 neocons

0.62

Activations Density 1.125%

phrases related to political, historical, and governmental entities

references to espionage and political intrigue

No Comments

No Known Activations