INDEX

Explanations

keys and numbers related to significant events or statistics

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

dates and years

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

 мәкал

-0.57

fjspx

-0.57

 שוליים

-0.54

StructEnd

-0.50

 ویکی‌پدی

-0.50

mybatisplus

-0.50

 Italijanski

-0.49

انيف

-0.47

 &___

-0.47

 kaarangay

-0.47

POSITIVE LOGITS

 meanwhile

0.44

 already

0.44

 déjà

0.43

didSet

0.40

 seguido

0.39

Investment

0.38

 Vázquez

0.38

 Viena

0.37

Intervention

0.37

BTW

0.37

Activations Density 0.045%

keys and numbers related to significant events or statistics

dates and years

No Comments

No Known Activations