INDEX

Explanations

elements related to legal or formal documentation

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

numbers and punctuation

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

 kasarigan

-1.12

出版年

-0.85

KURZBESCHREIBUNG

-0.84

expandindo

-0.83

Datuak

-0.80

principalTable

-0.79

 betweenstory

-0.77

SequentialGroup

-0.77

Portály

-0.76

OGND

-0.74

POSITIVE LOGITS

 special

0.30

la

0.30

 blood

0.30

ectoria

0.29

//

0.29

The

0.28

os

0.28

lo

0.28

clear

0.27

Activations Density 0.223%

elements related to legal or formal documentation

numbers and punctuation

No Comments

No Known Activations