© 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. 18246
    Prev
    Next
    INDEX
    Explanations

    resources

    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
    å¤Ħç½ļ
    -0.27
    ä¸įåıĬ
    -0.25
     Terrace
    -0.25
    lash
    -0.24
    æĥ©
    -0.24
    æijģ
    -0.24
    èĢĥ
    -0.24
    straint
    -0.24
    ä»·åĢ¼
    -0.23
    çļĦåħ´è¶£
    -0.23
    POSITIVE LOGITS
     related
    0.40
     center
    0.36
     libr
    0.36
    ä¸Ńå¿ĥ
    0.34
     centre
    0.33
     librarian
    0.32
     relative
    0.31
    center
    0.31
     Related
    0.31
     centers
    0.30
    Activations Density 0.961%

    No Known Activations