© 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. 18766
    Prev
    Next
    INDEX
    Explanations

    (

    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
     NC
    -0.57
    EC
    -0.57
    ec
    -0.56
    NC
    -0.55
    fc
    -0.55
    RC
    -0.55
     EC
    -0.55
    _rc
    -0.53
     SC
    -0.53
    rc
    -0.53
    POSITIVE LOGITS
    antasy
    0.33
     cla
    0.32
    çĶ¬
    0.26
    ually
    0.26
    cess
    0.25
    cla
    0.25
    ä¾ij
    0.25
    é¼¹
    0.25
    compan
    0.25
    cpp
    0.24
    Activations Density 10.168%

    No Known Activations