© 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. 32-TRANSCODER-HP
    7. 41980
    Prev
    Next
    INDEX
    Explanations

    Various situations

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    糧
    -0.28
    ANC
    -0.26
    ogle
    -0.25
    inges
    -0.24
    cod
    -0.24
    çĮ¿
    -0.24
    FRING
    -0.23
     wash
    -0.23
     ward
    -0.23
    æIJº
    -0.23
    POSITIVE LOGITS
    stricted
    0.29
     fused
    0.29
     acute
    0.28
    zeÅĽ
    0.28
     symp
    0.27
     distributed
    0.27
    çĶ°éĩİ
    0.27
     reversed
    0.27
    zego
    0.26
    æĢ¥æĢ§
    0.26
    Activations Density 0.004%

    No Known Activations