© 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. 18272
    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
    Not in Any Lists

    No Comments

    Negative Logits
    retty
    -0.28
     Silver
    -0.28
     instrumental
    -0.26
    Silver
    -0.25
    silver
    -0.25
    ä¸īæĺŁ
    -0.25
    rimp
    -0.25
    äºĶæĺŁ
    -0.24
    ä¸īæĿ¡
    -0.24
     Greece
    -0.24
    POSITIVE LOGITS
    åī§åľº
    0.28
    rase
    0.27
    å¡ijèĥ¶
    0.26
     Pal
    0.26
    æĪ·
    0.26
    åī§
    0.26
    ]+\
    0.26
    Pal
    0.25
    +len
    0.25
     juego
    0.25
    Activations Density 0.555%

    No Known Activations