© 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. 108148
    Prev
    Next
    INDEX
    Explanations

    species

    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
    èĥ½åĬĽåĴĮ
    -0.27
    oust
    -0.27
     Canter
    -0.25
    bove
    -0.24
    thon
    -0.24
     rospy
    -0.23
     sever
    -0.23
    绿èī²åıijå±ķ
    -0.23
    ä½Ĩä¸įéĻIJ
    -0.23
     bonuses
    -0.23
    POSITIVE LOGITS
    ewise
    0.30
    isse
    0.28
    wise
    0.25
    icked
    0.25
    åİ¢
    0.25
    æĪİ
    0.24
    麹
    0.23
    äºī夺
    0.23
    马åħĭ
    0.23
     kto
    0.23
    Activations Density 6.062%

    No Known Activations