© 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. 22-TRANSCODER-HP
    7. 97528
    Prev
    Next
    INDEX
    Explanations

    code, documents, math

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    enty
    -0.28
     """↵
    -0.27
    è¯
    -0.26
     """↵↵
    -0.25
     обеÑģп
    -0.24
    LOTS
    -0.24
    _kill
    -0.24
    WEEN
    -0.24
     GTX
    -0.24
    ahi
    -0.24
    POSITIVE LOGITS
    å·¥ä½ľæĸ¹æ¡Ī
    0.25
     scenario
    0.25
     linked
    0.24
    ä¸Ģä¸ĸ
    0.24
    æīĢ说
    0.24
     bounced
    0.23
    iams
    0.23
     sol
    0.23
    anax
    0.23
     contests
    0.23
    Activations Density 0.000%

    No Known Activations