© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 5104
    Prev
    Next
    INDEX
    Explanations

    Technical/legal context

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    imedia
    -0.28
    ngrx
    -0.26
    ichel
    -0.25
    è®Ń
    -0.25
    patch
    -0.25
     coincidence
    -0.25
    kn
    -0.25
    dn
    -0.25
    irie
    -0.24
     Hoe
    -0.24
    POSITIVE LOGITS
    误差
    0.27
    éĩĩ
    0.25
     Meter
    0.25
     Sixth
    0.24
    é£İ
    0.24
    _MM
    0.23
    èĮĦåŃIJ
    0.23
    URES
    0.23
     tasting
    0.23
    ajas
    0.23
    Activations Density 0.001%

    No Known Activations