© 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. 63106
    Prev
    Next
    INDEX
    Explanations

    data

    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
    longleftrightarrow
    -0.26
    大è¡Ĺ
    -0.25
    å±ķå¼Ģ
    -0.25
    scaled
    -0.25
    å¼Ģå±ķ
    -0.24
    'an
    -0.24
    ÄŁ
    -0.24
    ḷ
    -0.23
    _axis
    -0.23
     associate
    -0.23
    POSITIVE LOGITS
     hass
    0.28
    Canada
    0.26
    å¤§åĽ½
    0.25
    æĶ¾åĩº
    0.25
    å¹³åı°ä¸Ĭ
    0.25
    tparam
    0.25
    åĮħ裹
    0.25
    rä
    0.25
    tera
    0.24
     Canada
    0.24
    Activations Density 2.952%

    No Known Activations