© 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. 41640
    Prev
    Next
    INDEX
    Explanations

    mathematical formulas

    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
    ox
    -0.85
    OX
    -0.84
     ox
    -0.84
    ix
    -0.84
    IX
    -0.84
     PX
    -0.82
     Ox
    -0.80
    [ix
    -0.79
    NX
    -0.79
    æ°§åĮĸ
    -0.79
    POSITIVE LOGITS
    èĽ´
    0.27
    ogo
    0.27
    à¸Ĭม
    0.26
    ookie
    0.24
    æ¹Ħ
    0.24
    äºļ太
    0.24
    HashTable
    0.23
     tph
    0.23
     Engineer
    0.23
    eworthy
    0.22
    Activations Density 0.011%

    No Known Activations