INDEX

Explanations

terms and concepts related to technology and products

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

specific nouns

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

niſſe

-0.61

 betweenstory

-0.59

 صوتيه

-0.59

 queſto

-0.56

 يتيمه

-0.55

нгред

-0.54

setVerticalGroup

-0.54

 Geſch

-0.54

RegressionTest

-0.54

#+#

-0.53

POSITIVE LOGITS

AxisAlignment

0.33

 تضيفلها

0.33

 textStatus

0.33

 Exacts

0.31

PerformLayout

0.29

 sélectionnez

0.29

hå

0.29

 mate

0.28

 medži

0.28

 téléchargement

0.28

Activations Density 0.006%

terms and concepts related to technology and products

specific nouns

No Comments

No Known Activations