INDEX

Explanations

statistical terms and metrics related to model performance evaluation

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

evaluation metrics

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AutoresizingMask

-0.54

BufferException

-0.49

DockStyle

-0.48

)_/¯

-0.44

interopRequire

-0.44

']?>

-0.44

 IBOutlet

-0.41

ⓧ

-0.41

 initComponents

-0.40

 تضيفلها

-0.40

POSITIVE LOGITS

featureID

0.44

 ſever

0.42

 consideration

0.42

 classificação

0.41

 considération

0.40

 accuracy

0.40

 pesca

0.38

 conférences

0.38

 efficiency

0.38

 tonel

0.38

Activations Density 0.026%

statistical terms and metrics related to model performance evaluation

evaluation metrics

No Comments

No Known Activations