© 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. 109009
    Prev
    Next
    INDEX
    Explanations

    Scientific papers

    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
    .Dial
    -0.26
    amping
    -0.26
    æĿ¡
    -0.26
    详è§ģ
    -0.26
    _PTR
    -0.25
    踩
    -0.25
    Fil
    -0.24
    REW
    -0.24
     sprung
    -0.24
    æľīèĩªå·±çļĦ
    -0.24
    POSITIVE LOGITS
    -inst
    0.30
    ARB
    0.29
    icom
    0.27
    icient
    0.26
     Thur
    0.25
    ice
    0.25
    glas
    0.25
    Lint
    0.25
    nov
    0.24
    æŀľ
    0.24
    Activations Density 0.019%

    No Known Activations