© 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. 25-TRANSCODER-HP
    7. 109030
    Prev
    Next
    INDEX
    Explanations

    academic texts

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     bets
    -0.26
    ital
    -0.26
    æĹĹèĪ°åºĹ
    -0.25
    croft
    -0.24
     predator
    -0.24
    starter
    -0.24
     unintention
    -0.24
    istration
    -0.23
    =start
    -0.23
     predatory
    -0.23
    POSITIVE LOGITS
    adena
    0.31
    æł¸
    0.30
    awn
    0.28
    ryn
    0.27
    quam
    0.27
    avier
    0.27
     Density
    0.26
    æĥħ人
    0.26
     Letter
    0.25
    ç»Īç»ĵ
    0.25
    Activations Density 0.120%

    No Known Activations