© 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. 130672
    Prev
    Next
    INDEX
    Explanations

    be

    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
    æ´Ĺè¡£
    -0.28
     training
    -0.27
    ds
    -0.26
    d
    -0.26
    uly
    -0.26
     macro
    -0.26
    æĿı
    -0.24
     gan
    -0.24
    Macro
    -0.24
     Hudson
    -0.24
    POSITIVE LOGITS
    زا
    0.28
    -mf
    0.27
    å·¦
    0.25
     Getter
    0.25
    bose
    0.25
    æĽ´å¤ļä¿¡æģ¯
    0.24
    æĪĸå°Ĩ
    0.24
    ModelError
    0.23
    FormField
    0.23
    佶
    0.23
    Activations Density 0.023%

    No Known Activations