INDEX

Explanations

terms related to trophies and awards

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

awards for winners

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 OnInit

-0.42

synthesize

-0.41

ppuden

-0.39

 يتيمه

-0.38

ớ

-0.34

 Commencez

-0.34

ỏ

-0.34

Син

-0.33



-0.33

POSITIVE LOGITS

 trophy

0.63

الحياه

0.58

 Trophy

0.56

 hombro

0.56

 AssemblyCompany

0.56

 awards

0.56

 awarded

0.54

 trofeo

0.53

 AWARD

0.53

 trophies

0.53

Activations Density 0.003%

terms related to trophies and awards

awards for winners

No Comments

No Known Activations