© 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. 32-TRANSCODER-HP
    7. 41907
    Prev
    Next
    INDEX
    Explanations

    unusual text fragments

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĥħ
    -0.30
    ç»ıåķĨ
    -0.27
    ÑĤин
    -0.26
    æIJı
    -0.26
    rides
    -0.26
    æ´¥
    -0.26
    dap
    -0.25
    ersion
    -0.25
    注
    -0.24
    ALE
    -0.24
    POSITIVE LOGITS
    urat
    0.29
    ochond
    0.28
    omon
    0.28
    :hidden
    0.27
    ophil
    0.26
    ÙĩاÙħ
    0.25
    æ£Ĵ
    0.25
    ofi
    0.25
    joint
    0.24
     Lia
    0.24
    Activations Density 0.000%

    No Known Activations