© 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. 19482
    Prev
    Next
    INDEX
    Explanations

    com

    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
    ç»´å¥ĩ
    -0.31
    ularity
    -0.28
    åĽĽæĺ¯
    -0.26
    çģ«çĥ§
    -0.26
    &R
    -0.26
     mildly
    -0.26
    è£Ļ
    -0.25
     conj
    -0.25
    åħ±
    -0.25
    æľĪåºķ
    -0.25
    POSITIVE LOGITS
    åĻ¨
    0.30
    ERO
    0.29
    è¾Ļ
    0.28
    åıij
    0.27
    éric
    0.26
    æī¹åĪ¤
    0.25
    èł²
    0.25
    à¹Ģสมà¸Ń
    0.25
    era
    0.24
    arkin
    0.24
    Activations Density 0.005%

    No Known Activations