INDEX

Explanations

numerical data related to statistics

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

numbers followed by nouns or quantities

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

oredCriteria

-0.79

IUrlHelper

-0.72

GEBURTS

-0.72

 invokingState

-0.72

 ModelExpression

-0.71

 wireType

-0.71

httphttps

-0.68

AccessorTable

-0.64

 estekak

-0.63

 snippetHide

-0.63

POSITIVE LOGITS

 gegeven

0.35

 gezet

0.33

 buruk

0.33

 serez

0.32

 survivors

0.32

 continues

0.31

 elástica

0.31

 pulseras

0.30

 negatively

0.30

 renseignements

0.30

Activations Density 0.004%

numerical data related to statistics

numbers followed by nouns or quantities

No Comments

No Known Activations