© 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. 7313
    Prev
    Next
    INDEX
    Explanations

    non-standard characters

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

    The neuron fires strongly on tokens that begin new segments or clauses—i.e. the first word after a major boundary (start‐of‐text, sentence break, dash, etc.).

    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
    ​​​​
    0.53
    <strong>
    0.51
    ※
    0.47
    ↵↵
    0.46
    <h3>
    0.45
    ❖
    0.45
    ∎
    0.45
    鲸
    0.45
    ↵↵↵↵↵
    0.44
     एजेंसी
    0.44
    POSITIVE LOGITS
    लेली
    0.53
     presento
    0.50
     Murder
    0.50
     रहेको
    0.49
     habe
    0.48
     erschien
    0.48
     genannt
    0.47
    鍏
    0.47
    मध्ये
    0.46
     blevet
    0.46
    Activations Density 0.002%

    No Known Activations