© 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. 18551
    Prev
    Next
    INDEX
    Explanations

    continuity

    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
    çĮĽ
    -0.30
     Sampler
    -0.29
    æķ£
    -0.29
    褶
    -0.27
    éĺ³
    -0.27
    åįķä½į
    -0.27
     Coronavirus
    -0.25
    chos
    -0.25
    urch
    -0.25
    涨价
    -0.25
    POSITIVE LOGITS
     continuity
    0.42
     Contin
    0.32
    contin
    0.32
    继ç»Ń
    0.32
    Contin
    0.31
     continued
    0.31
     continuous
    0.31
     continua
    0.30
    _contin
    0.30
     contin
    0.30
    Activations Density 1.063%

    No Known Activations