© 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. 96152
    Prev
    Next
    INDEX
    Explanations

    Names and publications

    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
    åŁł
    -0.32
    硬
    -0.30
    éĢıæĺİ
    -0.29
    æ±ĩæĬ¥
    -0.28
     Cr
    -0.28
    lying
    -0.28
    licative
    -0.27
    liness
    -0.27
    alien
    -0.27
    å»ī
    -0.27
    POSITIVE LOGITS
    clause
    0.29
    æĹĹä¸ĭ
    0.27
     encour
    0.26
     clause
    0.26
    chief
    0.26
     fieldValue
    0.26
     chief
    0.25
     Plain
    0.25
    ocl
    0.25
    é¹ij
    0.24
    Activations Density 0.102%

    No Known Activations