© 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. 30-TRANSCODER-HP
    7. 96108
    Prev
    Next
    INDEX
    Explanations

    Code snippets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    maf
    -0.29
     shot
    -0.27
    bum
    -0.26
    è°¨
    -0.26
    ffe
    -0.25
    æĸĩä¸Ń
    -0.25
    jec
    -0.25
     Covent
    -0.25
     Affiliate
    -0.25
    èĢĥçĤ¹
    -0.24
    POSITIVE LOGITS
    æŀĦ
    0.29
    天河
    0.28
    UPDATED
    0.28
    èī²
    0.27
    ìĦ¤
    0.27
    .sep
    0.26
     bidi
    0.26
    anding
    0.25
    anca
    0.25
    uada
    0.25
    Activations Density 0.885%

    No Known Activations