© 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. 96099
    Prev
    Next
    INDEX
    Explanations

    acked

    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
    âĬ¿
    -0.27
    lete
    -0.27
    çŃīå¤ļ项
    -0.26
    ÑĥÑĩа
    -0.26
    LETE
    -0.25
    ongan
    -0.25
    缺
    -0.24
    omatic
    -0.24
     Markus
    -0.24
     EQ
    -0.24
    POSITIVE LOGITS
     those
    0.38
     the
    0.37
     their
    0.34
     with
    0.34
     again
    0.33
     them
    0.31
     its
    0.29
     definitely
    0.29
     still
    0.29
    åĨĴéĻ©
    0.28
    Activations Density 0.948%

    No Known Activations