© 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. 18466
    Prev
    Next
    INDEX
    Explanations

    patience

    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
    äºĶçĻ¾
    -0.27
    代表
    -0.26
     Prob
    -0.26
     Spicer
    -0.24
    é¢Ĥ
    -0.23
    çģ«èĬ±
    -0.23
    éĽ¾
    -0.23
    lia
    -0.23
    .viewmodel
    -0.23
    osal
    -0.22
    POSITIVE LOGITS
    pat
    1.26
     Pat
    1.16
     patient
    1.15
    Pat
    1.13
     pat
    1.11
    _pat
    1.08
    .pat
    1.02
    æĤ£èĢħ
    1.02
     PAT
    1.02
    patient
    1.02
    Activations Density 20.526%

    No Known Activations