© 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. 31-TRANSCODER-HP
    7. 18123
    Prev
    Next
    INDEX
    Explanations

    negative behavior or situations

    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
    -mort
    -0.28
    ç»ĵ
    -0.28
    çİ»
    -0.27
    æľºæ¢°
    -0.27
    毫
    -0.26
     kel
    -0.26
    åī½
    -0.26
    å¼ĢäºĨ
    -0.25
    严èĤĥ
    -0.25
    étr
    -0.25
    POSITIVE LOGITS
     Peak
    0.27
    ONSE
    0.26
     peak
    0.26
     tick
    0.25
    ctions
    0.25
    aji
    0.25
    BOX
    0.24
    èĪĮå°ĸ
    0.24
     displayName
    0.24
    ames
    0.24
    Activations Density 1.243%

    No Known Activations