© 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. 108938
    Prev
    Next
    INDEX
    Explanations

    try

    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
     Wilkinson
    -0.27
    ä¼ij
    -0.27
     congestion
    -0.25
    Å¡ÃŃ
    -0.25
    ETS
    -0.25
     domic
    -0.24
    ancements
    -0.24
    éĻį温
    -0.24
    east
    -0.23
    æŃ£åľ¨
    -0.23
    POSITIVE LOGITS
    ensely
    0.28
     ones
    0.27
    eni
    0.25
    è¾Ľ
    0.25
    eny
    0.25
    å¿į
    0.25
    _Release
    0.24
     Fate
    0.24
    çĽ
    0.23
     frequently
    0.23
    Activations Density 1.036%

    No Known Activations