© 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. 31-GEMMASCOPE-2-RES-16K
    7. 8209
    Prev
    Next
    INDEX
    Explanations

    challenges and critical distinction

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

    The neuron fires on the structured “header” lines and fact‐dense lead sentences of press‐style blurbs—e.g. album‐review headings (Artist – “Album” (Date):) or exhibition announcements—marking the start of those formal, information‐packed sections.

    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_31_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
     🥰
    1.16
    🥰
    1.14
     очень
    1.13
    🤣🤣
    1.08
     дуже
    1.04
     наверное
    1.02
     bardzo
    1.00
     VERY
    1.00
     labai
    0.98
     vrlo
    0.98
    POSITIVE LOGITS
     “[
    1.06
     yalnızca
    1.05
    “[
    1.04
    —
    0.98
    —”
    0.96
    ––
    0.94
    —“
    0.93
    ─
    0.87
    ُ
    0.84
     lackluster
    0.81
    Activations Density 0.436%

    No Known Activations