INDEX

Explanations

terminology related to testing and types in programming or data structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

andExpect

-0.37

 mewakili

-0.36

plaatst

-0.34

 kommunen

-0.33

 nahilalakip

-0.32

 alike

-0.32

IntoConstraints

-0.32

СТВ

-0.32

 lipat

-0.32

 representantes

-0.32

POSITIVE LOGITS

parsedMessage

0.62

 Biôgrafia

0.62

хьтан

0.60

layoutControl

0.58

webElementXpaths

0.57

 pinulongan

0.57

şört

0.57

DockStyle

0.56

 الرياضيه

0.55

 дописавши

0.54

Activations Density 0.006%

terminology related to testing and types in programming or data structures

No Comments

No Known Activations