© 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. 25-TRANSCODER-HP
    7. 108845
    Prev
    Next
    INDEX
    Explanations

    parts of words

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ek
    -0.33
    ec
    -0.31
    ef
    -0.31
    tal
    -0.29
    tur
    -0.29
    tu
    -0.28
    copy
    -0.28
    mem
    -0.27
    ram
    -0.27
    tam
    -0.27
    POSITIVE LOGITS
    ivid
    0.29
    favor
    0.28
    aisal
    0.27
    IBE
    0.27
    ivar
    0.26
    ingers
    0.26
    iversal
    0.26
    INGS
    0.25
    ifax
    0.25
    INGER
    0.25
    Activations Density 2.603%

    No Known Activations