© 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. Qwen3-4B
    5. 31-TRANSCODER-HP
    7. 18304
    Prev
    Next
    INDEX
    Explanations

    national

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    id
    -0.28
    coe
    -0.27
    RC
    -0.27
    à¹Ģà¸Ī
    -0.26
    jid
    -0.25
    rc
    -0.25
    icom
    -0.25
    нÑģÑĤ
    -0.24
    dx
    -0.24
     id
    -0.24
    POSITIVE LOGITS
    iard
    0.26
     mantle
    0.26
    å·´åŁº
    0.25
    oyer
    0.25
    aida
    0.24
    orsi
    0.24
    elpers
    0.24
    _intersection
    0.24
     Lama
    0.23
    elp
    0.23
    Activations Density 0.011%

    No Known Activations