© 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. 88391
    Prev
    Next
    INDEX
    Explanations

    hand

    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
    dy
    -0.27
    èĤĥ
    -0.25
    çŃĶ
    -0.25
     dumps
    -0.24
     Western
    -0.24
    lever
    -0.23
    INED
    -0.23
    asses
    -0.23
    cer
    -0.23
    gar
    -0.23
    POSITIVE LOGITS
    TYPO
    0.26
    建档ç«ĭ
    0.25
    gewater
    0.25
    èĢ³æľµ
    0.24
    åŀĭ
    0.24
    ConfigurationException
    0.24
    .bit
    0.24
     scal
    0.24
    _bits
    0.23
     itself
    0.23
    Activations Density 0.020%

    No Known Activations