© 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. 23573
    Prev
    Next
    INDEX
    Explanations

    4

    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
     LOT
    -0.26
     Geological
    -0.25
    hole
    -0.24
    çİij
    -0.24
    GeV
    -0.24
     Relief
    -0.24
     Replay
    -0.23
    ếp
    -0.23
     generals
    -0.23
    ica
    -0.23
    POSITIVE LOGITS
    åĮħ
    0.31
    alog
    0.28
     Conj
    0.28
    ration
    0.27
    é½IJ
    0.26
    uggested
    0.26
    æĿ¼
    0.26
    èĸªæ°´
    0.25
    梦è§ģ
    0.25
    lus
    0.25
    Activations Density 0.000%

    No Known Activations