© 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. 108376
    Prev
    Next
    INDEX
    Explanations

    mouse clicking

    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
    tsy
    -0.27
    eurs
    -0.26
    çĪ±ä½ł
    -0.24
     fifth
    -0.24
    omain
    -0.24
    å±ķå¼Ģ
    -0.24
    bai
    -0.24
    aniu
    -0.24
     Fifth
    -0.24
    è¿ĩçļĦ
    -0.23
    POSITIVE LOGITS
    æĵİ
    0.29
     penetrated
    0.28
     pregn
    0.27
    驱åĬ¨
    0.24
    TypeDef
    0.24
    åĩºåĬ¨
    0.23
    éĴı
    0.23
    éģ¿åŃķ
    0.23
    ä¸ĵèģĮ
    0.23
    ogui
    0.23
    Activations Density 0.145%

    No Known Activations