© 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. 19264
    Prev
    Next
    INDEX
    Explanations

    variations of "has"

    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
    箨
    -0.28
    ãģıãģªãĤĭ
    -0.27
    ä¸įåĨįæĺ¯
    -0.27
    lest
    -0.26
    ujÄħ
    -0.26
    orang
    -0.25
     become
    -0.24
     désormais
    -0.24
    [now
    -0.24
    ][(
    -0.23
    POSITIVE LOGITS
     has
    0.82
    Has
    0.72
    has
    0.71
     Has
    0.71
    	has
    0.69
    _has
    0.66
     hasn
    0.65
    .has
    0.63
     have
    0.62
     HAS
    0.62
    Activations Density 23.984%

    No Known Activations