© 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. 19602
    Prev
    Next
    INDEX
    Explanations

    profiling

    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
    æĴĤ
    -0.29
     plais
    -0.26
    é«ĺéĢŁåıijå±ķ
    -0.25
    rous
    -0.24
    客æłĪ
    -0.24
     apolog
    -0.24
    åľ°åŁŁ
    -0.23
     sche
    -0.23
    缮çļĦåľ°
    -0.23
    .Builder
    -0.23
    POSITIVE LOGITS
     profile
    0.85
    Profile
    0.74
     Profile
    0.74
    profile
    0.72
    _profile
    0.71
    -profile
    0.71
     profiles
    0.71
     PROFILE
    0.71
    PROFILE
    0.69
    .profile
    0.68
    Activations Density 2.940%

    No Known Activations