INDEX

Explanations

references to statements, questions, or definitions related to prior content and interactions

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

referring to something above

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

parsedMessage

-0.44

featureID

-0.41

Décès

-0.41

 appContext

-0.40

endphp

-0.39

 scolaires

-0.38

 extranjera

-0.37

 väx

-0.36

 Allociné

-0.36

ähteet

-0.36

POSITIVE LOGITS

fromnode

0.52

 صوتيه

0.49

httphttps

0.44

Карьера

0.43

 Chwiliwch

0.43

 autorytatywna

0.43

omiast

0.42

Tembelea

0.41

fore

0.41

 nahilalakip

0.41

Activations Density 0.346%

references to statements, questions, or definitions related to prior content and interactions

referring to something above

No Comments

No Known Activations