© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 7021
    Prev
    Next
    INDEX
    Explanations

    }

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

    The neuron strongly activates on numeric tokens (e.g. line numbers, code offsets, and other purely numeric annotations).

    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
    foo
    0.43
    Bark
    0.42
    できます
    0.41
    ():
    0.39
    to
    0.38
    ();
    0.38
    Dummy
    0.38
    𝘀
    0.38
    Lbl
    0.38
    that
    0.37
    POSITIVE LOGITS
     secretive
    0.42
     fiancée
    0.36
     itinerant
    0.35
     penchant
    0.35
     sorely
    0.34
     demokrat
    0.34
     contentious
    0.34
     preconceived
    0.34
     Pediatric
    0.33
     besieged
    0.33
    Activations Density 0.000%

    No Known Activations