© 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. 31-TRANSCODER-HP
    7. 18133
    Prev
    Next
    INDEX
    Explanations

    The

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rü
    -0.27
    æľį
    -0.24
    ferences
    -0.24
    apos
    -0.24
    eft
    -0.24
     definite
    -0.23
    igraph
    -0.23
    ible
    -0.23
     acre
    -0.23
    è¿ľæĸ¹
    -0.23
    POSITIVE LOGITS
     Cover
    0.27
    æĶ¶æķĽ
    0.26
    æģ½
    0.25
    .vertx
    0.24
    线
    0.24
    驱
    0.24
    å¯ĮåĬĽ
    0.24
    arez
    0.24
     subtitles
    0.24
    壤
    0.23
    Activations Density 35.895%

    No Known Activations