© 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. 31-TRANSCODER-HP
    7. 18144
    Prev
    Next
    INDEX
    Explanations

    Inv

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     att
    -0.29
    att
    -0.28
    伪
    -0.27
     ATM
    -0.27
    umpt
    -0.26
     rebound
    -0.26
     fest
    -0.26
     marked
    -0.25
     marc
    -0.25
    \Annotation
    -0.24
    POSITIVE LOGITS
    -inv
    0.35
    .Inv
    0.33
    Inv
    0.33
     Inv
    0.32
     INV
    0.32
    _inv
    0.29
    olved
    0.29
    lias
    0.28
    æ¶ī
    0.27
    çī½
    0.27
    Activations Density 0.141%

    No Known Activations