© 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. 97628
    Prev
    Next
    INDEX
    Explanations

    Names

    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
    çĮĿ
    -0.29
     R
    -0.28
    idel
    -0.27
    踵
    -0.27
    schemas
    -0.26
    uers
    -0.26
    entlich
    -0.26
    פ×Ķ
    -0.25
    çķĻä¸ĭ
    -0.25
     mild
    -0.24
    POSITIVE LOGITS
    è¿Ľåľº
    0.25
    olson
    0.25
    åĽłä¸ºå®ĥ
    0.25
    ÑĪен
    0.24
    AMS
    0.24
    Kernel
    0.24
    Ins
    0.24
     Mounted
    0.23
    Know
    0.23
     wherein
    0.23
    Activations Density 0.021%

    No Known Activations