© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 11883
    Prev
    Next
    INDEX
    Explanations

    phrases that express desire or intention

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

    Occurs before verbs like "win", "see", or "know"

    np_acts-logits-general · gemini-2.0-flash

    want to actions

    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
     lenker
    -0.66
    bondale
    -0.59
    ลิต
    -0.58
    UES
    -0.58
    :",
    -0.57
    aline
    -0.57
    vats
    -0.57
     ',
    -0.57
    "){
    -0.57
    toft
    -0.57
    POSITIVE LOGITS
     graag
    0.76
     liever
    0.71
    IActionResult
    0.67
    ագրություններ
    0.67
     know
    0.66
    orteur
    0.65
    paravant
    0.63
     entend
    0.62
     SUDOC
    0.62
     ILogger
    0.61
    Activations Density 0.111%

    No Known Activations