© 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. 54233
    Prev
    Next
    INDEX
    Explanations

    Programming and other languages

    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
     drifted
    -0.31
    ian
    -0.29
    asts
    -0.28
    ä¸Ģæľµ
    -0.25
    çĻ¾èĬ±
    -0.25
    acus
    -0.24
    arts
    -0.24
     CUSTOM
    -0.24
     trailed
    -0.24
    åħ¬ç§¯
    -0.24
    POSITIVE LOGITS
    ice
    0.30
    ä¸ĢåįĬ
    0.29
    GU
    0.27
    epy
    0.27
    PAL
    0.26
    uele
    0.25
    ico
    0.24
    essler
    0.24
    ica
    0.24
    çıŀ
    0.24
    Activations Density 0.029%

    No Known Activations