© 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. 31-TRANSCODER-HP
    7. 18917
    Prev
    Next
    INDEX
    Explanations

    play

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    enic
    -0.29
    unct
    -0.27
    åľ°å¸¦
    -0.26
    ç»ıå¼Ģ
    -0.25
    AutoSize
    -0.25
    éĥ½ä¸įçŁ¥éģĵ
    -0.25
    acket
    -0.25
    æݬ
    -0.24
    eries
    -0.24
    èIJĥ
    -0.24
    POSITIVE LOGITS
     play
    0.57
    Play
    0.54
    play
    0.52
     Play
    0.50
     PLAY
    0.49
    	play
    0.46
     plays
    0.45
    plays
    0.44
     played
    0.43
    çİ©
    0.43
    Activations Density 0.536%

    No Known Activations