© 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. 8157
    Prev
    Next
    INDEX
    Explanations

    research abstracts with "rat"

    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
    вин
    -0.31
    .identifier
    -0.27
    s
    -0.27
     hum
    -0.27
     unw
    -0.26
    ®
    -0.26
    çĵ
    -0.25
    è±Į
    -0.24
    åIJĪæ³ķæĿĥçĽĬ
    -0.24
    sns
    -0.24
    POSITIVE LOGITS
     YE
    0.28
    æĶ¹ç¼ĸ
    0.25
    ÓĢ
    0.25
    _spacing
    0.24
    -routing
    0.24
     corr
    0.24
    ISCO
    0.24
    CAPE
    0.23
    eks
    0.23
    ernetes
    0.23
    Activations Density 0.005%

    No Known Activations