© 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. 9024
    Prev
    Next
    INDEX
    Explanations

    back to

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

    The neuron is strongly triggered by instances of the word “back” (i.e. detecting uses of “back” indicating return or reversal).

    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
    j
    1.13
     éviter
    1.09
     atteindre
    1.05
    ్‌
    1.03
     revisar
    1.03
    jq
    1.01
     reemplazar
    1.00
     tenté
    1.00
    ii
    0.99
     marquées
    0.96
    POSITIVE LOGITS
     indietro
    1.30
    初心
    1.13
     एकदा
    1.11
    鹘
    1.11
     sharply
    1.05
    anlage
    1.03
    anlagen
    1.02
     vividly
    1.02
    クタ
    1.01
    dated
    0.98
    Activations Density 0.183%

    No Known Activations