© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 63756
    Prev
    Next
    INDEX
    Explanations

    healing and HP

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly activates on CamelCase programming identifiers—particularly enum‐ or struct‐style names (like “WantsToUseItem,” “ProvidesHealing,” “AreaOfEffect,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    🫁
    -0.90
    ňte
    -0.74
     Compliance
    -0.74
    dafx
    -0.74
    Compliance
    -0.73
    )-(
    -0.73
    沉
    -0.72
     lockdown
    -0.70
    Rigidbody
    -0.70
     WorldCat
    -0.70
    POSITIVE LOGITS
     healing
    3.31
     heal
    3.02
     heals
    2.69
    healing
    2.47
     healed
    2.39
    heal
    2.34
     Healing
    2.25
    Healing
    2.20
    Heal
    2.17
     Heal
    2.16
    Activations Density 0.012%

    No Known Activations