INDEX

Explanations

phrases related to inclusion or components of a system

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.76

下载附件

-0.71

IsContent

-0.70

 okuyayım

-0.66

ComponentModel

-0.65

 SafeMath

-0.65

 Signalez

-0.64

 ActiveRecord

-0.64

InstrumentedTest

-0.63

 Gunther

-0.60

POSITIVE LOGITS

terno

0.55

 Audiodateien

0.55

 kehadiran

0.55

complexContent

0.54

 existentes

0.53

homonymie

0.52

 kuin

0.52

ureusement

0.52

 kolei

0.51

utusan

0.51

Activations Density 0.016%

phrases related to inclusion or components of a system

No Comments

No Known Activations