© 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. 19123
    Prev
    Next
    INDEX
    Explanations

    database datas

    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
     instruments
    -0.32
    èģ²
    -0.31
    声
    -0.31
     mech
    -0.28
    jem
    -0.27
    org
    -0.27
    å·¥åħ·
    -0.26
    -org
    -0.26
     nec
    -0.26
     Instruments
    -0.26
    POSITIVE LOGITS
    _datas
    0.34
    database
    0.31
    æķ°æį®
    0.30
    .datab
    0.29
    datatable
    0.29
    Datas
    0.29
    datas
    0.28
    åĪĨæĶ¯
    0.28
     dab
    0.28
    .datas
    0.28
    Activations Density 0.239%

    No Known Activations