© 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. 25-TRANSCODER-HP
    7. 108107
    Prev
    Next
    INDEX
    Explanations

    Project plan

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    etro
    -0.30
    erde
    -0.26
    ipsis
    -0.26
    afia
    -0.26
    æĸ½å±ķ
    -0.26
    æĦıä¹īä¸ĬçļĦ
    -0.26
    æ°¿
    -0.25
    -striped
    -0.25
     benefited
    -0.24
    .returnValue
    -0.24
    POSITIVE LOGITS
     plan
    0.35
     Plan
    0.32
    plan
    0.32
    Plan
    0.32
     var
    0.29
    var
    0.29
     Projekt
    0.28
     initial
    0.28
    (plan
    0.27
     trial
    0.27
    Activations Density 1.200%

    No Known Activations