© 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. 32-TRANSCODER-HP
    7. 42049
    Prev
    Next
    INDEX
    Explanations

    Names starting with "Ph", saxophonists

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ÃŃt
    -0.31
     es
    -0.27
    马æĭī
    -0.27
     Smy
    -0.26
    -es
    -0.25
    imest
    -0.25
    ivities
    -0.24
    outline
    -0.24
    ÙĨاس
    -0.24
    BOR
    -0.23
    POSITIVE LOGITS
     Phantom
    0.35
    oph
    0.34
    ahr
    0.33
    Phill
    0.33
    Phil
    0.32
     Phil
    0.32
     Philip
    0.31
    åĵ²
    0.31
     Phill
    0.31
    phi
    0.31
    Activations Density 0.001%

    No Known Activations