© 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. 33-TRANSCODER-HP
    7. 53313
    Prev
    Next
    INDEX
    Explanations

    go

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åºı
    -0.29
    rollable
    -0.28
    ural
    -0.27
     sơ
    -0.27
    æijĨ
    -0.27
     {?}
    -0.26
    ady
    -0.25
    enen
    -0.25
    urement
    -0.25
    irtual
    -0.25
    POSITIVE LOGITS
    ighting
    0.27
    ms
    0.26
    MS
    0.26
    æĭĴç»Ŀ
    0.25
    åŁĭ
    0.25
    action
    0.25
     impuls
    0.24
    AGENT
    0.24
    raf
    0.24
    /action
    0.24
    Activations Density 0.006%

    No Known Activations