INDEX

Explanations

code-related terms, particularly those involving Android development

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

ⓧ

-0.41

 Carrillo

-0.37

 Barbosa

-0.36

 Egger

-0.35

AMPLES

-0.35

LEGGI

-0.34

 Nachbar

-0.34

 Jennings

-0.33

 znajdu

-0.33

 Delgado

-0.33

POSITIVE LOGITS

 AssemblyCulture

0.63

RenderAtEndOf

0.62

出版年

0.60

LookAnd

0.59

 AssemblyProduct

0.56

談社

0.56

 iprot

0.54

principalColumn

0.52

AnchorStyles

0.51

PreferredItem

0.50

Activations Density 0.008%

code-related terms, particularly those involving Android development

No Comments

No Known Activations