© 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. 8612
    Prev
    Next
    INDEX
    Explanations

    Col

    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
     deg
    -0.32
    ears
    -0.28
    htub
    -0.26
    æĥķ
    -0.26
    heatmap
    -0.25
    è¾±
    -0.25
    åį®
    -0.25
    çĽ¼
    -0.24
    为äºĨéģ¿åħį
    -0.24
     Pear
    -0.24
    POSITIVE LOGITS
    åŁ¹åħ»
    0.28
    ante
    0.26
    working
    0.26
    河西
    0.25
     reck
    0.25
    ãģĦãĤĭ
    0.25
    DMA
    0.24
    wick
    0.24
    ãĥĭãĥ³ãĤ°
    0.24
     DMA
    0.23
    Activations Density 0.158%

    No Known Activations