© 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. 18279
    Prev
    Next
    INDEX
    Explanations

    neglect

    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
    åºĦ
    -0.29
     doubly
    -0.27
    ara
    -0.27
    tim
    -0.26
    corn
    -0.26
    å·¡
    -0.26
    _UNIQUE
    -0.26
    交
    -0.25
    emails
    -0.25
    usk
    -0.25
    POSITIVE LOGITS
    忽è§Ĩ
    0.45
     neglect
    0.41
     Neg
    0.38
    忽çķ¥
    0.36
    neg
    0.36
     ignore
    0.33
    Neg
    0.33
     ignores
    0.33
    æĹłè§Ĩ
    0.33
     neglected
    0.32
    Activations Density 0.335%

    No Known Activations