© 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. 130249
    Prev
    Next
    INDEX
    Explanations

    Code and computer language

    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
    åıªè¦ģæľī
    -0.28
    åıªè¦ģæĺ¯
    -0.27
    uale
    -0.25
     trophy
    -0.25
    åĬ£
    -0.25
    æħ¢æħ¢çļĦ
    -0.24
     Levin
    -0.24
    åĩ¡æĺ¯
    -0.24
    ä¸ĢæŦ
    -0.23
     leven
    -0.23
    POSITIVE LOGITS
    w
    0.28
     itself
    0.27
    ations
    0.27
    ation
    0.27
    æĮijéĢī
    0.26
     chù
    0.26
     stretch
    0.25
     stretching
    0.25
    m
    0.25
    ç³»åĪĹ
    0.25
    Activations Density 0.001%

    No Known Activations