© 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. 18358
    Prev
    Next
    INDEX
    Explanations

    unusual characters

    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
    arden
    -0.29
    atisch
    -0.27
    ähr
    -0.27
    çļĦå¿ĥ
    -0.26
    ativ
    -0.26
    igr
    -0.26
    å¸Ī
    -0.25
    師
    -0.25
    american
    -0.25
     WH
    -0.25
    POSITIVE LOGITS
    ç»Ħç»ĩ
    0.28
    éļĪ
    0.25
    åłĥ
    0.24
    éĢ¼
    0.24
    Ŀ
    0.24
    é»į
    0.24
    uri
    0.23
    çīĸ
    0.23
    Jur
    0.23
    STRUCT
    0.23
    Activations Density 18.962%

    No Known Activations