INDEX

Explanations

references to specific car models and manufacturers

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

car models

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

Autoritní

-0.64

adaptiveStyles

-0.60

 disambiguazione

-0.58

protoimpl

-0.58

GEBURTS

-0.55

uestran

-0.53

әрмәләр

-0.50

UnsafeEnabled

-0.50

hyrchwyd

-0.49

webElementXpaths

-0.48

POSITIVE LOGITS

ashqai

0.50

 Passat

0.46

 Camry

0.44

 BaseActivity

0.44

 sedan

0.39

 Kombat

0.38

 Camaro

0.38

shund

0.38

 Accord

0.38

Hamlet

0.37

Activations Density 0.021%

references to specific car models and manufacturers

car models

No Comments

No Known Activations