© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 30-TRANSCODER-HP
    7. 96814
    Prev
    Next
    INDEX
    Explanations

    come

    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
    模åŀĭ
    -0.45
     model
    -0.39
    getModel
    -0.36
     Model
    -0.34
    .getModel
    -0.34
     models
    -0.34
     getModel
    -0.33
    _model
    -0.32
    	model
    -0.32
     modèle
    -0.32
    POSITIVE LOGITS
    Ĥ¬
    0.30
    ï¼Įé»ĺ认
    0.29
     integrated
    0.29
     defaults
    0.29
    Integrated
    0.28
    éħįä¸Ĭ
    0.28
    =default
    0.28
    pref
    0.27
    egrated
    0.27
    .DEFAULT
    0.27
    Activations Density 0.885%

    No Known Activations