© 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. 30-TRANSCODER-HP
    7. 95824
    Prev
    Next
    INDEX
    Explanations

    Smith

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     unre
    -0.27
    åĩºåħ¥
    -0.26
    userdata
    -0.26
     commercial
    -0.25
    eos
    -0.25
    оÑĢо
    -0.24
     UD
    -0.24
    ÅŁi
    -0.24
    UserData
    -0.24
    ees
    -0.24
    POSITIVE LOGITS
    son
    0.41
    enate
    0.33
    eson
    0.32
    sonian
    0.29
    .mods
    0.28
    bell
    0.28
    ä¿Ŀå®Ī
    0.28
    岫
    0.27
     Conserv
    0.27
    en
    0.26
    Activations Density 0.278%

    No Known Activations