© 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. 18682
    Prev
    Next
    INDEX
    Explanations

    Random unrelated text

    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
    èѦ
    -0.30
    å¹¹
    -0.30
     monet
    -0.27
    _KIND
    -0.25
     homogeneous
    -0.25
     cycl
    -0.25
    _DEPRECATED
    -0.25
    ogen
    -0.25
     Monroe
    -0.24
    éĴ§
    -0.24
    POSITIVE LOGITS
    æĽ¿
    0.25
     Fil
    0.24
    IDL
    0.24
    汶
    0.23
    ati
    0.23
    æ¡´
    0.23
    abay
    0.23
    OnInit
    0.23
     CrossRef
    0.23
    stack
    0.23
    Activations Density 0.893%

    No Known Activations