© 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. 95676
    Prev
    Next
    INDEX
    Explanations

    season

    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
     roman
    -0.27
     breakup
    -0.26
    rega
    -0.25
     rewriting
    -0.24
     bloom
    -0.24
    ATUS
    -0.24
    åĭĸ
    -0.24
    elin
    -0.24
     plung
    -0.24
    butt
    -0.24
    POSITIVE LOGITS
     season
    0.75
    åŃ£èĬĤ
    0.67
    åŃ£
    0.66
     Season
    0.65
     seasons
    0.60
    season
    0.59
    Season
    0.58
    _season
    0.58
     saison
    0.55
     seasonal
    0.52
    Activations Density 0.882%

    No Known Activations