© 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. 19471
    Prev
    Next
    INDEX
    Explanations

    7

    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
    Ly
    -0.28
    ظÙħ
    -0.27
    ÅĤu
    -0.27
     Ly
    -0.26
     гÑĢ
    -0.24
     purification
    -0.24
    osition
    -0.24
    ÑģÑĭ
    -0.24
    åĽ½éĻħåIJĪä½ľ
    -0.24
    æµĶ
    -0.24
    POSITIVE LOGITS
    isma
    0.27
    adan
    0.27
    罩
    0.27
     noreferrer
    0.26
    nia
    0.25
    év
    0.25
    æĻ¦
    0.25
    .firebaseio
    0.25
    éĢļéģĵ
    0.24
    åį·
    0.24
    Activations Density 0.002%

    No Known Activations