© 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. 12-TRANSCODER-HP
    7. 23279
    Prev
    Next
    INDEX
    Explanations

    web archives

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    chez
    -0.27
    åij»
    -0.26
    genes
    -0.25
    +-+-+-+-
    -0.25
    éķĬ
    -0.25
    ('__
    -0.24
    ixer
    -0.24
     entr
    -0.24
    該
    -0.24
     Fat
    -0.23
    POSITIVE LOGITS
     paramMap
    0.29
    -č↵
    0.28
    ditor
    0.28
     abortions
    0.28
    å¿Ļç¢Į
    0.25
    Ĥ¨
    0.24
    åĪĨ管
    0.24
     Transition
    0.23
    .xhtml
    0.23
     minlength
    0.23
    Activations Density 0.002%

    No Known Activations