INDEX

Explanations

terms related to programming or coding functions that manipulate fields or plugins

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

SELECT * means all columns

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

 صوتيه

-0.50

fromnode

-0.46

awtextra

-0.42

 للمعارف

-0.39

DropColumn

-0.39

 onOptions

-0.38

SBATCH

-0.38

ⓧ

-0.38

 legit

-0.37

 상세

-0.37

POSITIVE LOGITS

 oprot

0.50

 pacchetto

0.50

 pacote

0.50

 incluidos

0.47

 CURIAM

0.47

 automatique

0.47

 incluido

0.47

 gesamte

0.46

 ujednoznacz

0.46

 automática

0.46

Activations Density 0.225%

terms related to programming or coding functions that manipulate fields or plugins

SELECT * means all columns

No Comments

No Known Activations