INDEX

Explanations

negative indications regarding functionality and access

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

HandlerContext

-0.69

\{\\

-0.68

 disambiguazione

-0.64

bolista

-0.57

wachs

-0.57

IBOutlet

-0.56

}{@

-0.56

 autorytatywna

-0.54

homonymie

-0.54

__*/

-0.54

POSITIVE LOGITS

 stanovnika

0.51

ActiveRecord

0.49

 rapports

0.48

tonode

0.47

 usable

0.47

XmlRootElement

0.47

TouchEvent

0.47

entry

0.47

γγ

0.47

法

0.45

Activations Density 0.098%

negative indications regarding functionality and access

No Comments

No Known Activations