© 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. 33-TRANSCODER-HP
    7. 54389
    Prev
    Next
    INDEX
    Explanations

    sports ending in "ball"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    俱ä¹IJéĥ¨
    -0.37
    è¿Ļé¦ĸæŃĮ
    -0.32
     club
    -0.29
    åĦ¿åŃIJ
    -0.28
    arest
    -0.28
    ected
    -0.27
    çIJĥéĺŁ
    -0.27
    çıłæ±Ł
    -0.26
    æĥħåĨµè¿Ľè¡Į
    -0.26
    club
    -0.26
    POSITIVE LOGITS
     matches
    0.33
     match
    0.30
    _OPTS
    0.28
    Matches
    0.28
     Matches
    0.28
     Match
    0.27
    amarin
    0.26
    	Spring
    0.26
     authorities
    0.26
    matches
    0.25
    Activations Density 0.045%

    No Known Activations