INDEX

Explanations

code-related terms and structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 autorytatywna

-0.66

LookAnd

-0.66

 تضيفلها

-0.61

нгред

-0.54

 Nucle

-0.54

 Bulbs

-0.53

OGND

-0.53

iſten

-0.53

 يتيمه

-0.52

SDF

-0.52

POSITIVE LOGITS

 begge

0.45

Халык

0.43

 Könige

0.41

 berdua

0.39

 plástica

0.39

 juntos

0.36

 Aufla

0.36

 nabíz

0.35

 Ahnung

0.35

 keduanya

0.34

Activations Density 2.175%

code-related terms and structures

No Comments

No Known Activations