© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-1B
    3. 17-GEMMASCOPE-2-RES-16K
    4. 1869
    Prev
    Next
    INDEX
    Explanations

    holding things or concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-1b-pt/resid_post/layer_17_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
    אה
    0.94
    见的
    0.92
    പോലെ
    0.91
    DISABLED
    0.91
     unaccompanied
    0.89
    顯示
    0.89
    见
    0.89
    きの
    0.84
    েশনে
    0.82
     EQ
    0.81
    POSITIVE LOGITS
    hold
    1.63
    握
    1.39
     hold
    1.38
     onto
    1.31
    holds
    1.24
     held
    1.23
    held
    1.23
     holding
    1.22
    Hold
    1.22
    holding
    1.20
    Activations Density 0.183%

    No Known Activations