INDEX

Explanations

prepositions and phrases indicating relationships or connections in the text

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

followed by noun

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 pungkas

-0.40

 pungkasnya

-0.35

 perhaps

-0.31

 possibly

-0.31

 muerto

-0.30

 comenzará

-0.30

 kağıdı

-0.29

 şeyler

-0.28

 cielos

-0.28

 compartida

-0.28

POSITIVE LOGITS

 autorytatywna

0.82

Majority

0.77

Personendaten

0.72

cyklopedia

0.72

 utafitiHapana

0.71

majority

0.71

GEBURTSDATUM

0.71

 Мексичка

0.70

httphttps

0.70

 latest

0.69

Activations Density 0.465%

prepositions and phrases indicating relationships or connections in the text

followed by noun

No Comments

No Known Activations