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. 22-GEMMASCOPE-RES-16K
    6. 12818
    Prev
    Next
    INDEX
    Explanations

    specific locations or significant places within a narrative

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

    fictional worlds/locations

    np_acts-logits-general · gemini-2.0-flash

    fictional planets and worlds

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    HomeAsUpEnabled
    -0.67
    ScopeManager
    -0.67
    itinéraire
    -0.64
     estekak
    -0.64
     <=",
    -0.64
    parsedMessage
    -0.64
     Kannada
    -0.64
     surla
    -0.61
    yntaxException
    -0.61
     DialogInterface
    -0.61
    POSITIVE LOGITS
     fictional
    0.56
    humans
    0.52
     Goode
    0.50
     humains
    0.49
    談社
    0.49
    cire
    0.49
    unat
    0.47
    populated
    0.47
     Haven
    0.46
    stot
    0.46
    Activations Density 0.272%

    No Known Activations