© 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. 108175
    Prev
    Next
    INDEX
    Explanations

    Places and languages

    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
    inite
    -0.30
    æĹı
    -0.28
    raries
    -0.27
    å®ĩå®Ļ
    -0.27
    åħ¨èĥ½
    -0.26
    ynet
    -0.26
    our
    -0.26
    ä½°
    -0.26
    endants
    -0.25
    们
    -0.25
    POSITIVE LOGITS
     deposit
    0.27
    ascar
    0.27
    chy
    0.26
    sandbox
    0.26
    _TestCase
    0.25
    aches
    0.24
     disrupted
    0.24
    è¿Ļä¸ªåľ°åĽ¾
    0.23
    æľªæĪIJ
    0.23
     deposits
    0.23
    Activations Density 0.908%

    No Known Activations