INDEX

Explanations

references to legal or procedural contexts

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

 solches

-0.42

 Untersuch

-0.42

 engraçadas

-0.41

 Producción

-0.40

 vägen

-0.40

 cortesía

-0.40

nalités

-0.40

 Anordnung

-0.40

 klara

-0.39

 Handlung

-0.38

POSITIVE LOGITS

Erstellt

0.56

ArrowToggle

0.50

 tartalomajánló

0.43

PropertyChanging

0.42

capa

0.42

 MessageBoxIcon

0.42

<?

0.41

CAP

0.41

出版年

0.41

bil

0.40

Activations Density 0.154%

references to legal or procedural contexts

No Comments

No Known Activations