INDEX

Explanations

keywords associated with algorithms, particularly focusing on their properties and categories

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

 jButton

-0.47

 didik

-0.44

httphttps

-0.43

<?

-0.43

 PackageManager

-0.42

 foot

-0.42

 fread

-0.42

uyasha

-0.42

/**

-0.42

)}^{

-0.41

POSITIVE LOGITS

ALM

0.62

ALC

0.61

ALC

0.60

 Alva

0.58

 Lithuan

0.58

SBATCH

0.56

LLLL

0.55

Baldwin

0.55

ALK

0.54

Alta

0.54

Activations Density 0.744%

keywords associated with algorithms, particularly focusing on their properties and categories

No Comments

No Known Activations