© 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. 31-GEMMASCOPE-2-RES-16K
    7. 10220
    Prev
    Next
    INDEX
    Explanations

    crispy texture

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

    The neuron flags emphatic, attention-grabbing words—especially strong descriptors or “loud” tokens (e.g. upper-case or intensifying adjectives and nouns).

    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
     به‌عنوان
    0.86
    Х
    0.86
    ‒
    0.85
    һ
    0.84
    ሕ
    0.83
    ‑
    0.82
    ij
    0.80
    ኅ
    0.79
    а
    0.79
    ḥ
    0.75
    POSITIVE LOGITS
     গুলো
    1.36
     bbq
    1.27
     thing
    1.21
     horrible
    1.21
     cosas
    1.19
     crappy
    1.16
     sorta
    1.14
     coisas
    1.13
     ব্যাবস্থা
    1.13
     soooo
    1.13
    Activations Density 0.974%

    No Known Activations