© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 30-TRANSCODER-HP
    7. 95248
    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_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     topic
    -0.31
     question
    -0.27
     Dwight
    -0.25
    åı¦ä¸ĢåįĬ
    -0.25
    宣è¨Ģ
    -0.24
    æ¶Ī
    -0.24
    è´¨çĸij
    -0.23
     subject
    -0.23
     sujet
    -0.23
    pest
    -0.23
    POSITIVE LOGITS
     (_,
    0.29
    èī¯å¿ĥ
    0.29
    apse
    0.27
     Busy
    0.27
    åIJį声
    0.27
    æĭ¿åĪ°äºĨ
    0.25
    -black
    0.25
    åŃ³
    0.24
     busy
    0.24
    ç®Ģ
    0.24
    Activations Density 0.003%

    No Known Activations