© 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. 22-TRANSCODER-HP
    7. 98375
    Prev
    Next
    INDEX
    Explanations

    Places

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åĮħ
    -0.28
     Bundle
    -0.27
    ened
    -0.27
    -metadata
    -0.27
    _PRESENT
    -0.26
    omers
    -0.26
    (layers
    -0.26
     %+
    -0.25
    ething
    -0.25
    æĸĩåĮĸä¼łæĴŃ
    -0.24
    POSITIVE LOGITS
    اض
    0.29
    çļĦæľĭåıĭ
    0.27
    ä»ħ为
    0.26
     Artificial
    0.26
    uda
    0.26
    éĢł
    0.25
    Unit
    0.25
    리를
    0.24
     Measures
    0.24
    å¼
    0.24
    Activations Density 0.033%

    No Known Activations