© 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. 32-TRANSCODER-HP
    7. 42850
    Prev
    Next
    INDEX
    Explanations

    wellbores and hydrocarbon production

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ray
    -0.33
    æ¢ģ
    -0.31
     beam
    -0.31
     Ray
    -0.29
    ray
    -0.28
    -validate
    -0.28
     Beam
    -0.28
    Ray
    -0.28
    PE
    -0.27
    æĿĥçĽĬ
    -0.27
    POSITIVE LOGITS
    cade
    0.28
    minster
    0.28
    çĸı
    0.26
    æ°´åºĵ
    0.26
    inear
    0.26
    éĢĤ
    0.26
    uspendLayout
    0.25
    ZONE
    0.25
    dge
    0.25
    ä¸İæŶ
    0.24
    Activations Density 0.031%

    No Known Activations