© 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. 33-TRANSCODER-HP
    7. 52986
    Prev
    Next
    INDEX
    Explanations

    small amount

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lug
    -0.32
    gie
    -0.28
    lico
    -0.27
    £¼
    -0.26
     uncompressed
    -0.25
    GIN
    -0.25
    lator
    -0.25
     kilomet
    -0.25
    麼
    -0.24
    éĽĨåĽ¢èĤ¡ä»½
    -0.24
    POSITIVE LOGITS
     less
    0.72
     few
    0.71
     fewer
    0.69
     Few
    0.65
    few
    0.63
     fully
    0.59
    less
    0.58
     Less
    0.56
    Few
    0.55
    _less
    0.53
    Activations Density 0.040%

    No Known Activations