© 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. 25-TRANSCODER-HP
    7. 108356
    Prev
    Next
    INDEX
    Explanations

    Code and markup language

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    为æŃ¢
    -0.43
    èĢĮè¨Ģ
    -0.41
    éĤ£æł·
    -0.39
    æĿ¥è¯´
    -0.36
    åľ¨åĨħçļĦ
    -0.34
    ä¸Ģæł·
    -0.33
    为ä¾ĭ
    -0.33
    æĥħåĨµæĿ¥çľĭ
    -0.33
    为åĩĨ
    -0.33
    #endregion
    -0.32
    POSITIVE LOGITS
     freder
    0.27
     massages
    0.25
     Fuse
    0.25
     Kurd
    0.24
    č↵		č↵
    0.24
    precation
    0.24
    ivalence
    0.23
    éªĪ
    0.23
    å¾Ĵ
    0.23
     hott
    0.23
    Activations Density 1.581%

    No Known Activations