© 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. 111773
    Prev
    Next
    INDEX
    Explanations

    code and legal classes/units

    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
     UIResponder
    -0.28
    廳
    -0.27
     conform
    -0.26
     cram
    -0.25
     politics
    -0.25
    agan
    -0.24
     aio
    -0.24
    olicies
    -0.24
    _ue
    -0.24
     eins
    -0.24
    POSITIVE LOGITS
    å·¥
    0.29
    éķ¿
    0.28
    亡
    0.27
    å·²æľī
    0.27
    å·²
    0.26
    nb
    0.26
    åıªè¦ģ
    0.26
    leh
    0.25
    订
    0.25
     éĶ
    0.25
    Activations Density 0.016%

    No Known Activations