© 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. 35-TRANSCODER-HP
    7. 38946
    Prev
    Next
    INDEX
    Explanations

    ges

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æī¿åĮħ
    -0.26
     undocumented
    -0.26
    (Collision
    -0.25
     endregion
    -0.25
    WithIdentifier
    -0.25
    (Intent
    -0.24
    _CN
    -0.24
    author
    -0.24
    -ci
    -0.24
    éŃħåĬĽ
    -0.24
    POSITIVE LOGITS
    ges
    0.36
    ose
    0.33
    oses
    0.31
    еÑģÑĮ
    0.30
    upe
    0.29
    ses
    0.29
    xes
    0.29
    mate
    0.28
    tes
    0.28
    ront
    0.28
    Activations Density 0.847%

    No Known Activations