© 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. 30-TRANSCODER-HP
    7. 96147
    Prev
    Next
    INDEX
    Explanations

    clouds

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    /+
    -0.28
    Latch
    -0.25
    åºŁ
    -0.25
    oji
    -0.24
    DAO
    -0.24
    滥
    -0.24
    æ²¥
    -0.24
    des
    -0.24
    ä¸ĩåħĥ以ä¸Ĭ
    -0.24
    eldon
    -0.24
    POSITIVE LOGITS
    _cloud
    0.30
    .cloud
    0.30
     clouds
    0.28
    åĽ±
    0.28
    cloud
    0.27
    å¤Ļ
    0.27
     sleeping
    0.26
    éĽ²
    0.26
     predicting
    0.25
     cloud
    0.25
    Activations Density 0.101%

    No Known Activations