INDEX

Explanations

terms related to programming logic and structure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

 незавершена

-0.83

 kasarigan

-0.79

 autorytatywna

-0.71

Jeografia

-0.64

nemies

-0.60

 يتيمه

-0.60

 дописавши

-0.60

 Numerade

-0.59

Портали

-0.55

+#+

-0.54

POSITIVE LOGITS

de

0.46

del

0.40

 kysy

0.39

da

0.38

Personensuche

0.37

bahnen

0.34

 kesä

0.33

 насељу

0.33

Répondre

0.33

 della

0.32

Activations Density 0.110%

terms related to programming logic and structure

No Comments

No Known Activations