© 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. 32-TRANSCODER-HP
    7. 41388
    Prev
    Next
    INDEX
    Explanations

    enc words

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    建议
    -0.38
    åĬĿ
    -0.32
    æݨèįIJ
    -0.32
    æıIJè®®
    -0.31
    æĮĩ导
    -0.31
    迫使
    -0.31
    è¦ģæ±Ĥ
    -0.30
    æıIJéĨĴ
    -0.30
    åIJ¸
    -0.30
    suggest
    -0.30
    POSITIVE LOGITS
     Enc
    1.02
    enc
    0.96
    Enc
    0.95
     enc
    0.95
    _enc
    0.83
    .enc
    0.78
    ENC
    0.76
     ENC
    0.75
    (enc
    0.74
     enclosure
    0.69
    Activations Density 0.005%

    No Known Activations