© 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. 30-TRANSCODER-HP
    7. 95910
    Prev
    Next
    INDEX
    Explanations

    many different things

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    è¾ĥ大
    -0.32
    ibilities
    -0.29
     dope
    -0.29
    è¾ĥ大çļĦ
    -0.29
    holes
    -0.28
    -rise
    -0.26
    ãĥ¼ãĥ³
    -0.26
    åĬ©çIJĨ
    -0.25
    åĴ«
    -0.25
     enorm
    -0.25
    POSITIVE LOGITS
     class
    0.42
     part
    0.40
    class
    0.39
    éĥ¨åĪĨ
    0.39
     pur
    0.39
    ç±»
    0.37
    滥
    0.33
    part
    0.31
     classe
    0.31
    Part
    0.30
    Activations Density 0.106%

    No Known Activations