© 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. 6714
    Prev
    Next
    INDEX
    Explanations

    without

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

    The neuron flags content-bearing words (nouns, verbs, adjectives, etc.) while ignoring common function words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    words or phrases indicating method, manner, or means of performing an action (often following prepositions like "by", "using", "from", or "through").

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    prepositional phrases that express means/method, accompaniment, location, movement, or other relational links within a sentence.

    oai_token-act-pair · gpt-5Triggered 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
     để
    0.79
     لإ
    0.77
    用来
    0.71
     เพื่อ
    0.70
     そして
    0.66
     للإ
    0.66
    对于
    0.65
     quản
    0.63
     对于
    0.63
    สำหรับการ
    0.63
    POSITIVE LOGITS
     without
    2.03
    without
    1.92
     WITHOUT
    1.91
     без
    1.84
     tanpa
    1.81
     melalui
    1.75
     באמצעות
    1.75
     senza
    1.74
     pomocí
    1.74
     using
    1.73
    Activations Density 1.733%

    No Known Activations