INDEX

Explanations

numerical values and their formats relevant to measurements and concentrations

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

quantity per area

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

 propOrder

-0.75

ⓧ

-0.68

 dévelo

-0.66

 ब्रेकडाउन

-0.64

SequentialGroup

-0.63

ientras

-0.63

كويكب

-0.63

 pouvoit

-0.63

 plufieurs

-0.60

 mauva

-0.59

POSITIVE LOGITS

 area

0.49

0.48

 Area

0.48

Area

0.47

area

0.46

 areas

0.40

endregion

0.40

 Areas

0.40

areas

0.39

aliana

0.36

Activations Density 0.107%

numerical values and their formats relevant to measurements and concentrations

quantity per area

No Comments

No Known Activations