© 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. 33-TRANSCODER-HP
    7. 54202
    Prev
    Next
    INDEX
    Explanations

    ://

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    locked
    -0.28
     mcc
    -0.28
    åĮ»çĸĹåį«çĶŁ
    -0.25
    è¿IJåĬ¨
    -0.25
     attacking
    -0.24
    ç¨Ģ
    -0.24
    èĬ±æł·
    -0.23
    assi
    -0.23
    dens
    -0.23
     Hell
    -0.23
    POSITIVE LOGITS
    çªĿ
    0.29
    æĸĵ
    0.25
     shel
    0.25
    æĩĭ
    0.24
    otine
    0.24
    ksam
    0.24
    ob
    0.24
    ueva
    0.24
    put
    0.24
     customs
    0.24
    Activations Density 0.004%

    No Known Activations