© 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. 43828
    Prev
    Next
    INDEX
    Explanations

    to

    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
    æĭ±
    -0.27
     маÑģÑĤеÑĢ
    -0.24
     effects
    -0.24
     propri
    -0.24
     effect
    -0.24
    ç»ķ
    -0.24
    éĢģåĩº
    -0.24
    effects
    -0.24
    Expression
    -0.23
    erver
    -0.23
    POSITIVE LOGITS
     test
    0.50
    æµĭè¯ķ
    0.46
     exam
    0.46
     tested
    0.44
     examined
    0.43
     tests
    0.43
     look
    0.42
     determined
    0.41
    èĢĥæŁ¥
    0.41
     teste
    0.41
    Activations Density 0.052%

    No Known Activations