© 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. 96397
    Prev
    Next
    INDEX
    Explanations

    international topics

    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
    åĪ°
    -0.36
     ahead
    -0.30
     Ahead
    -0.30
     heavens
    -0.29
    -forward
    -0.28
    åĬ¡
    -0.28
    -ahead
    -0.27
    åĨįåĪ°
    -0.27
    -away
    -0.27
    _FWD
    -0.26
    POSITIVE LOGITS
     into
    0.66
    into
    0.61
     Into
    0.60
    Into
    0.56
    _into
    0.52
     INTO
    0.46
     vÃło
    0.39
    è¿Ľåħ¥
    0.38
    .into
    0.38
    éĢ²åħ¥
    0.32
    Activations Density 0.899%

    No Known Activations