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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 16-GEMMASCOPE-RES-16K
    6. 5239
    Prev
    Next
    INDEX
    Explanations

    references to towns and similar geographical entities

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

    "town" and "towns"

    np_acts-logits-general · gemini-2.0-flash

    town locations like "X of Y"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Réponses
    -0.70
    fromCharCode
    -0.58
    requestData
    -0.56
    aust
    -0.56
     lượng
    -0.54
     final
    -0.53
    shtml
    -0.53
     ağı
    -0.51
     Koa
    -0.51
    ijnen
    -0.51
    POSITIVE LOGITS
     towns
    1.11
     town
    1.09
    towns
    1.03
     TOWN
    1.01
     CommonModule
    0.93
    Town
    0.93
    Towns
    0.93
     Town
    0.91
     Towns
    0.91
    WebServlet
    0.90
    Activations Density 0.040%

    No Known Activations