INDEX

Explanations

scientific and statistical terminology, particularly related to significance and p-values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Tikang

-0.40

Autoritní

-0.40

 OnInit

-0.33

єра

-0.33

 themselves

-0.32

下り

-0.31

 gọn

-0.30

 abstracta

-0.29

 سپاه

-0.29

POSITIVE LOGITS

'**

0.64

"*"

0.64

'*'

0.63

 asterisk

0.63

 étoile

0.62

('*',

0.60

 bintang

0.59

("*

0.59

('*

0.58

">*</

0.58

Activations Density 0.142%

scientific and statistical terminology, particularly related to significance and p-values

No Comments

No Known Activations