INDEX

Explanations

terms related to spectral analysis and spectral indices

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

spectral properties

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 business

-0.47

 démoc

-0.44

 mother

-0.43

home

-0.43

 männis

-0.43

 Gegenteil

-0.41

margin

-0.41

 home

-0.41

 money

-0.40

 politician

-0.40

POSITIVE LOGITS

 spectral

2.02

 Spectral

1.95

Spectral

1.88

spectral

1.80

 Spectra

0.79

\{\\

0.73

 спек

0.72

 Spect

0.72

 spectra

0.71

 SPECT

0.71

Activations Density 0.011%

terms related to spectral analysis and spectral indices

spectral properties

No Comments

No Known Activations