© 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. 31-TRANSCODER-HP
    7. 18777
    Prev
    Next
    INDEX
    Explanations

    gram

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    riad
    -0.30
    ç¾Į
    -0.28
    vit
    -0.26
    /kg
    -0.26
    æĸ¤
    -0.25
     Counseling
    -0.24
    æ³ī
    -0.24
    çıĤ
    -0.24
    åIJĬ
    -0.24
    åĶĩ
    -0.24
    POSITIVE LOGITS
     g
    0.60
    g
    0.60
     Gram
    0.46
     г
    0.46
    gram
    0.46
    ï½ĩ
    0.45
     gram
    0.43
    -g
    0.42
    gr
    0.42
     gr
    0.42
    Activations Density 46.539%

    No Known Activations