© 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. 27-TRANSCODER-HP
    7. 9723
    Prev
    Next
    INDEX
    Explanations

    solid

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    åİŁåĪĻ
    -0.28
    akter
    -0.27
    aming
    -0.27
     land
    -0.27
    åĺī
    -0.26
    mony
    -0.26
    éģĵ
    -0.25
    iscard
    -0.25
     fiat
    -0.25
    GA
    -0.24
    POSITIVE LOGITS
     slashing
    0.28
     rendez
    0.28
    getPath
    0.27
     Algorithm
    0.27
     suicides
    0.26
    èĢģå¸ĪçļĦ
    0.26
    å¹³åĿĩæ¯ı
    0.26
    ä¸įæĩĪ
    0.25
    nings
    0.25
    æĶ¾
    0.25
    Activations Density 0.799%

    No Known Activations