© 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. 24-GEMMASCOPE-2-RES-16K
    7. 13571
    Prev
    Next
    INDEX
    Explanations

    point in time

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

    the neuron detects words immediately preceding numeric tokens (i.e. it lights up on the token that comes just before a number).

    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
     इसलिये
    0.73
    neurs
    0.70
     તેમાં
    0.67
     termina
    0.66
    缕
    0.66
    许多
    0.65
    蕾
    0.65
    🦒
    0.63
     તેથી
    0.63
    煖
    0.63
    POSITIVE LOGITS
    us
    0.88
    ia
    0.82
    ä
    0.82
    ﮑ
    0.81
     আলোকে
    0.80
     Можно
    0.80
    ﻨ
    0.78
    ஸ்த
    0.77
    iam
    0.77
     Voz
    0.77
    Activations Density 0.000%

    No Known Activations