© 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. 20-TRANSCODER-HP
    7. 108262
    Prev
    Next
    INDEX
    Explanations

    Locations, dates, origins

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     rac
    -0.27
     invested
    -0.27
    ControlEvents
    -0.26
    å¢ĥ
    -0.25
     lun
    -0.25
    echa
    -0.24
     invest
    -0.24
     ec
    -0.24
    ieme
    -0.24
    routeProvider
    -0.24
    POSITIVE LOGITS
    åĴĮåıijå±ķ
    0.29
    ongan
    0.26
    UGH
    0.26
    ore
    0.25
    åħ´èµ·
    0.25
    æ£Ģçĸ«
    0.25
     felt
    0.25
    çľīæ¯Ľ
    0.25
    æĦĽ
    0.24
    ÑħваÑĤ
    0.24
    Activations Density 16.266%

    No Known Activations