© 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. 108108
    Prev
    Next
    INDEX
    Explanations

    rap

    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
    gether
    -0.28
     Extra
    -0.28
    park
    -0.27
    @class
    -0.26
    xford
    -0.26
    åĵģ
    -0.26
    æŀ«
    -0.26
    flare
    -0.24
    -zA
    -0.24
     extra
    -0.24
    POSITIVE LOGITS
    achu
    0.29
    enstein
    0.26
    emachine
    0.26
    anela
    0.26
    IMA
    0.26
    érie
    0.25
    çªĹå¸ĺ
    0.25
    slide
    0.25
    éķ¿çĽ¸
    0.24
    iscal
    0.24
    Activations Density 0.099%

    No Known Activations