© 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. 20-TRANSCODER-HP
    7. 108212
    Prev
    Next
    INDEX
    Explanations

    ten

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    éĥijéĩį
    -0.27
    hips
    -0.25
    ivot
    -0.24
     vita
    -0.24
    缸åıį
    -0.24
    ua
    -0.24
    å¸Ĥ
    -0.24
     venir
    -0.23
    uae
    -0.23
    ivia
    -0.23
    POSITIVE LOGITS
    lte
    0.28
    IEWS
    0.26
    =subprocess
    0.25
    åĩºå¢ĥ
    0.24
     scout
    0.24
    ying
    0.24
    lage
    0.24
    æĻļé¥Ń
    0.23
    ally
    0.23
    .OUT
    0.23
    Activations Density 16.710%

    No Known Activations