© 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. 30-TRANSCODER-HP
    7. 95915
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    .setType
    -0.28
    æłī
    -0.27
     Boeh
    -0.26
    Ñĥз
    -0.26
     Golf
    -0.24
    æĸ°éĹ»ä¸Ńå¿ĥ
    -0.24
     Exercise
    -0.24
     Adapt
    -0.24
     AppConfig
    -0.23
    LANGUAGE
    -0.23
    POSITIVE LOGITS
     dine
    0.26
     pick
    0.26
     valid
    0.26
     dye
    0.25
     invalid
    0.25
    sing
    0.25
    æĺ¯ä¸Ģéĥ¨
    0.25
    ami
    0.25
    engers
    0.25
    oken
    0.25
    Activations Density 0.894%

    No Known Activations

    This feature has no known activations.