© 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. 18940
    Prev
    Next
    INDEX
    Explanations

    Verification

    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
    æ£ĢæŁ¥
    -0.29
    ãĥģãĤ§ãĥĥãĤ¯
    -0.28
     Check
    -0.27
    æª¢æŁ¥
    -0.27
    å´Ĩ
    -0.26
     check
    -0.26
    inalg
    -0.25
     sac
    -0.25
    .Check
    -0.24
    çĽijçĿ£æ£ĢæŁ¥
    -0.24
    POSITIVE LOGITS
    éªĮ
    0.73
    éªĮè¯ģ
    0.70
    é©Ĺ
    0.68
    ver
    0.68
     verifies
    0.65
     verification
    0.65
     verify
    0.64
    VER
    0.63
     validate
    0.62
    -ver
    0.61
    Activations Density 27.710%

    No Known Activations