© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 11270
    Prev
    Next
    INDEX
    Explanations

    local details

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

    The neuron strongly activates on tokens that are numeric parameter values or measurements (i.e. numbers and their associated units) in technical/instructional text.

    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
    ↵
    1.02
    s
    1.01
    AR
    0.98
    RA
    0.93
    ség
    0.87
    lab
    0.86
    BA
    0.84
    e
    0.82
    Pr
    0.80
    t
    0.80
    POSITIVE LOGITS
     USDC
    0.88
    ঙালি
    0.80
     rigorously
    0.78
     cryptographic
    0.78
     memperoleh
    0.77
     overstated
    0.77
     Crosby
    0.75
     किरदार
    0.75
     najważ
    0.75
     untersucht
    0.75
    Activations Density 0.000%

    No Known Activations