© 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. 14555
    Prev
    Next
    INDEX
    Explanations

    identifying similarity

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

    The main thing this neuron does is detect comparative language—words and phrases that set up relative comparisons (e.g. “costlier,” “more,” “similar,” “other … models,” “have … their own take”).

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

    comparisons or discussions of alternatives, competitors, or similar options.

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

    comparative or contrastive statements that discuss alternatives, options, or similar items (e.g., comparing one thing to others).

    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
    เดิม
    0.68
    ல்கள்
    0.63
     kembali
    0.62
     terakhir
    0.61
    従来の
    0.56
     ഏറ്റവും
    0.56
     pertama
    0.55
     সর্বশেষ
    0.55
     изначально
    0.55
    従来
    0.55
    POSITIVE LOGITS
     similarly
    1.39
     similaire
    1.36
     similar
    1.34
     similaires
    1.27
    相同
    1.25
     ተመሳሳይ
    1.25
     비슷한
    1.20
     비슷
    1.20
    类似的
    1.19
     Similarly
    1.18
    Activations Density 0.370%

    No Known Activations