© 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. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 8239
    Prev
    Next
    INDEX
    Explanations

    numbers and code

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

    The neuron is responding to tokens at a particular absolute position in the input (i.e. it fires on tokens that occur around a fixed location in the document).

    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
    ó
    0.88
    기
    0.87
    म
    0.85
    २
    0.74
    <0x92>
    0.71
    ında
    0.71
    м
    0.68
    ாஹ
    0.68
     universality
    0.66
    ]}$.
    0.66
    POSITIVE LOGITS
     Dodson
    0.97
     Lied
    0.84
     Kline
    0.79
     izquierdo
    0.79
     Eintrag
    0.78
     milhões
    0.77
     alami
    0.77
    搭配
    0.76
     Flickr
    0.75
     అధికారు
    0.75
    Activations Density 0.000%

    No Known Activations