INDEX

Explanations

elements related to structured data or API definitions

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

httphttps

-0.34

 wachten

-0.33

 виправивши

-0.32

 prisonniers

-0.32

znamu

-0.31

 miniaturka

-0.31

 næste

-0.30

 pungkas

-0.30

 prisoners

-0.29

ztat

-0.29

POSITIVE LOGITS

 AssemblyCulture

0.69

__(/*!

0.66

rawDesc

0.61

Autoritní

0.57

MLLoader

0.52

 Spot

0.52

 ExecuteAsync

0.51

 препратки

0.50

setof

0.50

EClass

0.50

Activations Density 0.189%

elements related to structured data or API definitions

No Comments

No Known Activations