© 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. 31-TRANSCODER-HP
    7. 18269
    Prev
    Next
    INDEX
    Explanations

    scientific studies

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    DAC
    -0.30
    åįĬ个æľĪ
    -0.29
    udes
    -0.28
    æªĹ
    -0.27
    积æŀģä½ľçĶ¨
    -0.26
    roph
    -0.26
     salv
    -0.25
    ÅĽcie
    -0.25
    åĩ¿
    -0.25
     imm
    -0.25
    POSITIVE LOGITS
    asaki
    0.26
    Spacing
    0.25
    Sig
    0.25
    åľ¨é¦Ļ港
    0.25
     naming
    0.25
    åľ¨å¸Ĥåľºä¸Ĭ
    0.25
     reordered
    0.25
     Sit
    0.25
    åĸĩ
    0.25
    okit
    0.24
    Activations Density 3.485%

    No Known Activations