INDEX

Explanations

terms related to popular products and trends

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Kerja

-0.41

.*")]

-0.37

 للاسماء

-0.33

 Merdeka

-0.31

 합니다

-0.31

gnore

-0.31

クリル

-0.31

.")]

-0.29

 ✭✭

-0.29

 فريبيس

-0.29

POSITIVE LOGITS

KEYCODE

0.55

GraphicsUnit

0.54

 many

0.54

Ӕ

0.52

AISSEE

0.51

writeFieldEnd

0.51

нгред

0.51

 Audiodateien

0.50

 estekak

0.49

+#+

0.49

Activations Density 0.040%

terms related to popular products and trends

most popular

No Comments

No Known Activations