INDEX

Explanations

years from the 1990s

oai_token-act-pair · gpt-3.5-turbo

references to specific decades, particularly those related to the 1990s

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

-1.08

MAP

-0.68

åĩ

-0.66

Anonymous

-0.65

 Unch

-0.64

 Anonymous

-0.64

 Piper

-0.63

é¾įå

-0.62

JP

-0.62

 Trin

-0.61

POSITIVE LOGITS

 respectively

0.79

ixties

0.79

esses

0.76

oons

0.76

 thereof

0.74

 thereafter

0.74

storms

0.71

ulia

0.70

ines

0.70

inement

0.70

Activations Density 0.081%

years from the 1990s

references to specific decades, particularly those related to the 1990s

No Comments

No Known Activations