© 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. 109077
    Prev
    Next
    INDEX
    Explanations

    anonymity

    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
    atie
    -0.25
    ç»ı纪人
    -0.24
    ä¹ı
    -0.24
     insult
    -0.24
    (Container
    -0.24
     colon
    -0.24
    composite
    -0.24
    容åĻ¨
    -0.24
    åĮĹéĥ¨
    -0.24
    ensible
    -0.23
    POSITIVE LOGITS
    éļIJ
    0.42
    éļIJèĶ½
    0.40
     anonymity
    0.39
    éļIJèĹı
    0.39
     hidden
    0.39
    éļIJ身
    0.37
    (hidden
    0.36
    éļ±
    0.36
     Hide
    0.35
    :hidden
    0.35
    Activations Density 0.904%

    No Known Activations