© 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. 22830
    Prev
    Next
    INDEX
    Explanations

    suffix

    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
    kop
    -0.31
    æķ´é½IJ
    -0.29
     Belt
    -0.27
    ása
    -0.27
    ynchron
    -0.25
    ffect
    -0.25
    iked
    -0.25
    åĪĨæķ£
    -0.24
    escription
    -0.24
     breakup
    -0.24
    POSITIVE LOGITS
    æĥĩ
    0.26
     pedig
    0.26
    æĽ²çº¿
    0.25
     Dw
    0.25
     Doll
    0.24
    èµĦ产è¯Ħä¼°
    0.24
    åı®åĺ±
    0.24
    |`↵
    0.24
    Fo
    0.24
     proxies
    0.24
    Activations Density 0.007%

    No Known Activations