© 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. 31-TRANSCODER-HP
    7. 18429
    Prev
    Next
    INDEX
    Explanations

    liver

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    issues
    -0.28
    _streams
    -0.25
     issue
    -0.25
     Streams
    -0.25
    éĵİ
    -0.25
    issue
    -0.24
     pint
    -0.24
    åIJĪãģĦ
    -0.24
    ç·
    -0.24
     Issues
    -0.24
    POSITIVE LOGITS
    jem
    0.27
    -pro
    0.26
     CARE
    0.26
    Care
    0.25
     prés
    0.25
    Fans
    0.25
    æ¿Ģæ´»
    0.25
    cÃŃ
    0.24
    æŀ£
    0.24
    ourg
    0.24
    Activations Density 0.816%

    No Known Activations