Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 16149
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to towns and cities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    appearing before names of populated places

    np_acts-logits-general · gemini-2.0-flash

    cities and towns, largest or main

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     estekak
    -0.77
    XmlAccessorType
    -0.67
     resourceCulture
    -0.64
     JpaRepository
    -0.63
    XmlAccessType
    -0.62
     <<<<<<<<<<<<<<
    -0.62
    AlterField
    -0.59
     AliExpress
    -0.59
     Lanzarote
    -0.58
     חיצוניים
    -0.57
    POSITIVE LOGITS
     city
    1.76
     cities
    1.62
     town
    1.55
     città
    1.38
     towns
    1.37
     ciudad
    1.31
    city
    1.27
     cidade
    1.26
     ciudades
    1.25
     villes
    1.20
    Activations Density 0.211%

    No Known Activations