© 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. 19-TRANSCODER-HP
    7. 7850
    Prev
    Next
    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     knock
    -0.27
    年轻人
    -0.27
    èĥĮåIJİ
    -0.26
    :B
    -0.25
    éķ¿æķĪæľºåĪ¶
    -0.25
    uba
    -0.25
    ständ
    -0.24
    ä¹ħ
    -0.23
    -label
    -0.23
    ä¸ī人
    -0.23
    POSITIVE LOGITS
    çļĦåľ°æĸ¹
    0.25
    çłĶ究éĻ¢
    0.25
    алÑĮнÑĭÑħ
    0.25
     pers
    0.24
    çļĦæīĢæľī
    0.24
    ATAR
    0.24
     ÑĢÑĥб
    0.23
    ioned
    0.23
    EGIN
    0.23
    onomic
    0.23
    Activations Density 0.531%

    No Known Activations