© 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. 27-TRANSCODER-HP
    7. 8887
    Prev
    Next
    INDEX
    Explanations

    role-playing games

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    olib
    -0.28
    rna
    -0.28
    erval
    -0.26
    zac
    -0.26
    ç«ĭ
    -0.25
     artikel
    -0.24
    åĩ¿
    -0.24
    çĶ·å£«
    -0.24
    LOAT
    -0.24
     Cou
    -0.24
    POSITIVE LOGITS
     GM
    0.41
    GM
    0.35
     campaign
    0.35
     referee
    0.33
     GMC
    0.30
     PCs
    0.30
    æ¡ĮåŃIJä¸Ĭ
    0.30
     campaigns
    0.30
     ruling
    0.29
     gm
    0.29
    Activations Density 0.012%

    No Known Activations