© 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. 112355
    Prev
    Next
    INDEX
    Explanations

    Code, errors, file paths

    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
    ervlet
    -0.27
    æĹłè§Ĩ
    -0.25
    éĢĶå¾Ħ
    -0.24
    LOCKS
    -0.24
    ates
    -0.23
    _perc
    -0.23
    MeshPro
    -0.23
    ä¼ijæģ¯
    -0.23
     command
    -0.23
    ctor
    -0.23
    POSITIVE LOGITS
    uder
    0.26
    gon
    0.26
     overlay
    0.26
    å¢Ļä¸Ĭ
    0.25
    _src
    0.25
     src
    0.24
    -motion
    0.24
    ä¸įä½ı
    0.24
    /unit
    0.24
    -auto
    0.24
    Activations Density 0.005%

    No Known Activations