© 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. 27-TRANSCODER-HP
    7. 8300
    Prev
    Next
    INDEX
    Explanations

    internet content and random languages

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lp
    -0.29
    lli
    -0.27
    mens
    -0.27
     hours
    -0.25
    解
    -0.25
    éĢłç¦ı
    -0.25
    æ¯ĶåĪ©
    -0.24
     global
    -0.24
    èĤŁ
    -0.24
    缩
    -0.24
    POSITIVE LOGITS
    æĢ»çIJĨ
    0.26
    strap
    0.25
     zipper
    0.24
     zakÅĤad
    0.24
    .pivot
    0.23
    serializer
    0.23
    åŁŃ
    0.23
    è¿IJ
    0.23
    _qual
    0.23
    avax
    0.23
    Activations Density 0.003%

    No Known Activations