© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-4-31B
    5. 30-RES-MATRYOSHKA-131K
    7. 119655
    Prev
    Next
    INDEX
    Explanations

    source code comments

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/gemma-4-saes/gemma-4-31b
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    其中的
    -0.08
    фект
    -0.06
     any
    -0.06
    either
    -0.06
    compassing
    -0.06
    umphed
    -0.06
    Dynamics
    -0.06
     quisquam
    -0.06
    Без
    -0.05
    Without
    -0.05
    POSITIVE LOGITS
     constantly
    0.07
     hastily
    0.07
     för
    0.07
     பயிற்சிய
    0.07
    레
    0.06
     overly
    0.06
     일부
    0.06
    IsNot
    0.06
    によっては
    0.06
     برخی
    0.06
    Activations Density 0.014%

    No Known Activations