© 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. 12-TRANSCODER-HP
    7. 23389
    Prev
    Next
    INDEX
    Explanations

    negative values

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     TORT
    -0.29
    itu
    -0.28
    åīįè¿Ľ
    -0.28
    ogs
    -0.28
    orts
    -0.28
    arp
    -0.26
    burgh
    -0.25
    èĢģ太太
    -0.24
     smb
    -0.24
    interval
    -0.24
    POSITIVE LOGITS
    ализ
    0.30
    serter
    0.26
    ä¸Ģæŀļ
    0.26
    sumer
    0.26
    IENT
    0.25
    æ½Ń
    0.25
    tery
    0.25
    duit
    0.24
    tape
    0.24
     taÅŁÄ±
    0.24
    Activations Density 0.001%

    No Known Activations