INDEX

Explanations

punctuation marks and sentence end indicators

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

šanās

-0.47

نجليز

-0.46

tojas

-0.45

 noël

-0.44

 choisissez

-0.44

luß

-0.43

 meille

-0.43

 maravilloso

-0.43

 ddelweddau

-0.42

 magique

-0.41

POSITIVE LOGITS

 EconPapers

0.56

+#+#

0.55

下载附件

0.51

PerformLayout

0.49

 snippetHide

0.48

تقاوى

0.48

ponses

0.47

MLLoader

0.47

 fubject

0.46

帖最后由

0.46

Activations Density 0.348%

punctuation marks and sentence end indicators

No Comments

No Known Activations