© 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. 19346
    Prev
    Next
    INDEX
    Explanations

    too/so

    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
    æĽ´é«ĺçļĦ
    -0.40
    æĽ´å¤§çļĦ
    -0.36
    æĽ´å¥½çļĦ
    -0.34
     bigger
    -0.33
     quantity
    -0.32
    better
    -0.31
    éĩı
    -0.31
    æĽ´ä½İ
    -0.29
     amount
    -0.29
     enough
    -0.29
    POSITIVE LOGITS
    å°½åı¯èĥ½
    0.33
     relatively
    0.27
    ially
    0.27
    ä¸įè§ģ
    0.26
    å°½åĬĽ
    0.25
    èĿ¶
    0.24
     Forge
    0.24
    orarily
    0.24
    å°½éĩı
    0.24
    udi
    0.24
    Activations Density 10.496%

    No Known Activations