© 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. 8054
    Prev
    Next
    INDEX
    Explanations

    Technical/programming context

    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
    quat
    -0.28
    jan
    -0.26
    reverse
    -0.25
    .gradle
    -0.24
    astr
    -0.24
    óry
    -0.24
    edo
    -0.23
     hãy
    -0.23
    åħ¬åı¸å°Ĩ
    -0.23
    itm
    -0.23
    POSITIVE LOGITS
    ica
    0.31
    icina
    0.28
     ext
    0.25
    ëĵľ
    0.25
    icare
    0.25
    ICA
    0.25
    ow
    0.25
    å®ļäºĨ
    0.25
    æĬ
    0.25
    ic
    0.25
    Activations Density 1.073%

    No Known Activations