© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 8808
    Prev
    Next
    INDEX
    Explanations

    start of things

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

    This neuron isn’t picking out any particular word or meaning—instead it spikes on whatever token falls at a specific fixed position in the input sequence.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    𝘪
    0.75
     chiropractor
    0.75
    Snackbar
    0.73
     lubricant
    0.72
     एंटीऑ
    0.71
    నిక
    0.71
     folic
    0.70
     antibiotic
    0.70
     fastener
    0.70
    ⵅ
    0.69
    POSITIVE LOGITS
    entrum
    0.79
     jah
    0.74
    inizin
    0.74
    igare
    0.70
    endim
    0.70
    leri
    0.69
    endish
    0.68
     jih
    0.68
     túi
    0.67
     אם
    0.66
    Activations Density 0.000%

    No Known Activations