© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 7273
    Prev
    Next
    INDEX
    Explanations

    Time periods/durations

    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
    bpp
    -0.29
    iele
    -0.27
    羣çĪ±
    -0.27
     INCIDENT
    -0.26
    uren
    -0.25
    溢价
    -0.25
    éĢļ常æĺ¯
    -0.24
     bef
    -0.24
    èĤ¯å®ļæĺ¯
    -0.24
    chia
    -0.24
    POSITIVE LOGITS
     sampling
    0.27
    åĦĴ
    0.27
     KeyCode
    0.26
     attraction
    0.26
    _gr
    0.24
     rigid
    0.24
    .mixin
    0.24
    CRT
    0.23
     deadline
    0.23
     Gr
    0.23
    Activations Density 0.006%

    No Known Activations