© 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. 95287
    Prev
    Next
    INDEX
    Explanations

    for

    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
     dist
    -0.27
    éĺ´
    -0.27
    èĵ¼
    -0.26
    vin
    -0.26
    dist
    -0.25
    ıntı
    -0.24
    ean
    -0.24
    ={$
    -0.23
    IfExists
    -0.23
     distances
    -0.23
    POSITIVE LOGITS
    æĽ´æľī
    0.35
     better
    0.35
     improved
    0.32
    better
    0.31
    Ending
    0.30
    æĽ´èĥ½
    0.29
    æĽ´åħ·
    0.28
     stronger
    0.28
    æĽ´å¼º
    0.28
     more
    0.28
    Activations Density 0.887%

    No Known Activations