© 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. 13047
    Prev
    Next
    INDEX
    Explanations

    comparisons and conclusions

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

    The neuron detects speaker-turn indicators in transcripts (i.e. names or labels introducing a new speaker’s line).

    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
     บริ
    0.70
    ัฒ
    0.68
     พ
    0.66
    atau
    0.64
    URAL
    0.63
    OPTIM
    0.63
     xây
    0.63
     シンプル
    0.62
     人気
    0.62
     lĩnh
    0.62
    POSITIVE LOGITS
     contradict
    0.80
     precluded
    0.79
     perempt
    0.77
    也不能
    0.76
     etiam
    0.74
     contradicted
    0.74
     निष्कर्ष
    0.72
    ❓
    0.71
    为什么
    0.71
     miscon
    0.71
    Activations Density 0.063%

    No Known Activations