INDEX

Explanations

percentage values in the text

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

percentage sign

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

тистика

-0.31

pol

-0.31

原始内容存档于

-0.29

both

-0.29

 gobern

-0.29

ciąż

-0.28

 ciertamente

-0.28

 تعدى

-0.28

Nadie

-0.28

Quali

-0.28

POSITIVE LOGITS

MLLoader

0.81

 المعيارى

0.71

 Walkover

0.70

fjspx

0.69

 ProtoMessage

0.68

LookAnd

0.66

SBATCH

0.63

='%

0.62

 للاسماء

0.62

Activations Density 0.029%

percentage values in the text

percentage sign

No Comments

No Known Activations