© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 31-TRANSCODER-HP
    7. 19728
    Prev
    Next
    INDEX
    Explanations

    the

    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
    Negative Logits
    èĵ¬
    -0.31
    egend
    -0.30
     sqr
    -0.30
    çĥŃçĪ±
    -0.29
    emer
    -0.28
    emat
    -0.28
     legend
    -0.27
    æģĤ
    -0.26
    portun
    -0.26
    -runner
    -0.26
    POSITIVE LOGITS
    istine
    0.26
     recovery
    0.26
    imeter
    0.26
    纡
    0.26
    oper
    0.25
    _REST
    0.24
    apers
    0.24
    ycl
    0.24
    GED
    0.24
     starting
    0.24
    Activations Density 0.311%

    No Known Activations