INDEX

Explanations

terms related to championship titles and sports accolades

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

winning championships and titles

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

RegressionTest

-0.87

 Infórmanos

-0.69

Personensuche

-0.65

 resourceCulture

-0.65

 disambiguazione

-0.62

fjspx

-0.62

 صوتيه

-0.61

 nahilalakip

-0.60

 deſſen

-0.60

OrBuilder

-0.60

POSITIVE LOGITS

 defeating

0.60

 beating

0.57

 twice

0.56

 victory

0.55

 undefeated

0.52

🏆

0.51

 winning

0.51

 victories

0.50

 triumph

0.49

twice

0.47

Activations Density 0.040%

terms related to championship titles and sports accolades

winning championships and titles

No Comments

No Known Activations