© 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. 9582
    Prev
    Next
    INDEX
    Explanations

    explicit sexual content vs abstract

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

    This neuron fires strongly on occurrences of the words “abstract” or “concrete.”

    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
    dock
    0.60
    itesse
    0.60
    ρν
    0.59
    𝚄
    0.59
    jährige
    0.58
    膳
    0.58
    鐗
    0.58
    履歴
    0.58
     கொண்டிருக்கும்
    0.57
    šenja
    0.57
    POSITIVE LOGITS
     абстра
    0.96
     abstract
    0.84
    abstract
    0.83
    Abstract
    0.82
     Abstract
    0.81
     abstra
    0.81
    Concrete
    0.80
     concrete
    0.78
    抽象
    0.78
     Concrete
    0.76
    Activations Density 0.083%

    No Known Activations