INDEX

Explanations

copyright-related terms and licenses

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

copyright and license notices

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

exitRule

-0.71

RegressionTest

-0.68

 HasFactory

-0.64



-0.61

RenderAtEndOf

-0.59

новниш

-0.55

+#+#

-0.54

UnusedPrivate

-0.53

Spoljašnje

-0.52

 ModelRenderer

-0.52

POSITIVE LOGITS

 juridiques

0.38

 padrões

0.32

 eût

0.32

 étoit

0.32

Искәрмәләр

0.32

Билгалдахарш

0.31

Fordítás

0.31

 voelen

0.31

 étoient

0.30

 poema

0.29

Activations Density 0.029%

copyright-related terms and licenses

copyright and license notices

No Comments

No Known Activations