© 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. 87762
    Prev
    Next
    INDEX
    Explanations

    Scientific discourse and evaluation

    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
    çłģ
    -0.29
    çļĦè¿ĽæŃ¥
    -0.28
    +B
    -0.26
    碼
    -0.25
    +k
    -0.25
     scor
    -0.24
    į°
    -0.24
    ëĪĪ
    -0.24
     stamped
    -0.24
    è¯Ń
    -0.23
    POSITIVE LOGITS
    åĪ³
    0.28
    MethodInfo
    0.25
    aker
    0.25
     infiltration
    0.25
    oref
    0.24
    cxx
    0.24
    ä»İæł¹æľ¬
    0.24
    &&!
    0.24
    unds
    0.24
    ivec
    0.24
    Activations Density 0.007%

    No Known Activations