© 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. 10392
    Prev
    Next
    INDEX
    Explanations

    as if / as she / as he

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

    This neuron is acting as a document‐position detector, firing more strongly the farther into the text it reaches.

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

    the word "as" used as a conjunction or preposition to introduce comparisons, temporal relationships, or simultaneous actions.

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

    subordinating conjunctions linking clauses, especially the form that introduces adverbial time/manner/cause clauses.

    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
     having
    0.88
     needing
    0.87
    knowing
    0.87
    having
    0.85
    snacks
    0.80
    event
    0.78
    supporting
    0.78
    很少
    0.78
    ের
    0.77
    saving
    0.77
    POSITIVE LOGITS
     negara
    0.80
     Rockville
    0.78
     మీరు
    0.75
     bạn
    0.73
     você
    0.73
    leduj
    0.72
     serta
    0.71
     scuole
    0.70
     σου
    0.69
     họ
    0.68
    Activations Density 0.040%

    No Known Activations