© 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. 32-TRANSCODER-HP
    7. 42468
    Prev
    Next
    INDEX
    Explanations

    primary, secondary

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    çİīçŁ³
    -0.28
    åĬ²
    -0.24
    æŀ¶
    -0.24
    æ°IJ
    -0.23
    表çİ°
    -0.23
    çľ¯
    -0.22
    ILITY
    -0.22
    qi
    -0.22
    urt
    -0.22
     pod
    -0.22
    POSITIVE LOGITS
     primary
    1.32
     Primary
    1.25
    primary
    1.20
    Primary
    1.19
     secondary
    1.18
     Secondary
    1.10
    secondary
    1.10
    (primary
    1.06
    Secondary
    1.06
    PRIMARY
    1.02
    Activations Density 0.001%

    No Known Activations