© 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. 12-TRANSCODER-HP
    7. 22318
    Prev
    Next
    INDEX
    Explanations

    Military units

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     roku
    -0.28
    campo
    -0.28
    ÑīÑĥ
    -0.28
    htable
    -0.27
     perse
    -0.25
    enza
    -0.25
    ceive
    -0.25
    à¸Ķวà¸ĩ
    -0.24
     Webb
    -0.23
    èµĦæºIJæķ´åIJĪ
    -0.23
    POSITIVE LOGITS
    иг
    0.28
    iac
    0.27
    pir
    0.26
    éĹ´æİ¥
    0.26
    å¼Ģæ°´
    0.26
    ilig
    0.25
    's
    0.25
    ItemList
    0.25
     multip
    0.24
    ä¸įæĪIJ
    0.24
    Activations Density 0.001%

    No Known Activations