© 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. 14-TRANSCODER-HP
    7. 41192
    Prev
    Next
    INDEX
    Explanations

    Asking questions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_14.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.14.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    egra
    -0.29
    ürü
    -0.29
     Iranians
    -0.29
     \"{
    -0.28
    å¡ĺ
    -0.26
    fäll
    -0.25
    çĶŁæĹ¥
    -0.25
     sm
    -0.25
    erca
    -0.25
     Egyptians
    -0.24
    POSITIVE LOGITS
    ÑĥÑĩ
    0.28
    tam
    0.26
     suit
    0.26
     past
    0.26
    å¥Ĺ
    0.25
    æĥħèĬĤ
    0.25
    _past
    0.25
    大åѦ
    0.25
    etri
    0.25
    AMI
    0.25
    Activations Density 0.064%

    No Known Activations