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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 6292
    Prev
    Next
    INDEX
    Explanations

    the neuron is looking for phrases related to physical contact and positioning, particularly involving the ground or surface.

    oai_token-act-pair · gemini-2.0-flash

    the ground

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     الحره
    -0.56
     ujednoznacz
    -0.52
    المناصب
    -0.50
    autoreleasepool
    -0.49
    DotNetBar
    -0.49
    RenderAtEndOf
    -0.48
     esternos
    -0.46
    angor
    -0.46
     withal
    -0.46
    iligt
    -0.46
    POSITIVE LOGITS
     ground
    1.96
    ground
    1.69
     earth
    1.64
    Ground
    1.63
     Ground
    1.59
     GROUND
    1.58
    GROUND
    1.47
     Earth
    1.45
     soil
    1.40
    earth
    1.32
    Activations Density 1.191%

    No Known Activations