© 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. 27-TRANSCODER-HP
    7. 57669
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Does
    -0.92
     does
    -0.85
    Does
    -0.83
     Are
    -0.81
    Are
    -0.75
     Is
    -0.69
     Did
    -0.68
    .Are
    -0.67
     Can
    -0.66
    does
    -0.66
    POSITIVE LOGITS
    èmes
    0.24
    DON
    0.24
    çĶŃ
    0.24
    Whether
    0.23
    åħļå§Ķ书记
    0.23
    noinspection
    0.23
     OpCode
    0.23
    whether
    0.23
    !=-
    0.23
    tdown
    0.23
    Activations Density 0.012%

    No Known Activations