INDEX

Explanations

causal relationships or origins

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate causes or origins of phenomena

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

kees

-0.63

©¶æ¥µ

-0.62

kids

-0.61

bats

-0.58

nam

-0.58

zon

-0.58

itiz

-0.58

abases

-0.57

hello

-0.57

orders

-0.56

POSITIVE LOGITS

 sheer

0.65

abal

0.65

 necessity

0.63

âĵĺ

0.62

Sov

0.61

stanbul

0.56

uting

0.56

 extracting

0.55

 knowing

0.55

 deliberate

0.54

Activations Density 0.347%

causal relationships or origins

phrases that indicate causes or origins of phenomena

No Comments

No Known Activations