© 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. 18-TRANSCODER-HP
    7. 89755
    Prev
    Next
    INDEX
    Explanations

    Code, academic, documentation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     spons
    -0.28
    æ©IJ
    -0.26
    kte
    -0.26
    abilit
    -0.26
    çĭĢ
    -0.25
     thừa
    -0.25
    åħ±åIJĮä½ĵ
    -0.25
    liga
    -0.25
    éħĬ
    -0.25
     Probe
    -0.24
    POSITIVE LOGITS
    è¾Ĺ转
    0.28
    ças
    0.26
     later
    0.25
    æĩĭ
    0.25
     nature
    0.25
     square
    0.24
    0.24
    spm
    0.24
     squares
    0.24
    åĪ©åĻ¨
    0.24
    Activations Density 45.424%

    No Known Activations