© 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. 25-TRANSCODER-HP
    7. 108260
    Prev
    Next
    INDEX
    Explanations

    civil

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æŶåĪ»
    -0.25
    paralle
    -0.25
     ALOG
    -0.25
    elog
    -0.25
     pylint
    -0.25
    ç¬ijç¬ij
    -0.25
    åºĨ幸
    -0.24
    hall
    -0.24
     dint
    -0.24
    Mer
    -0.24
    POSITIVE LOGITS
     Ñģог
    0.29
    isin
    0.28
    scheme
    0.25
    åĭĥåĭĥ
    0.24
    ly
    0.24
    bow
    0.23
    ìļĶ
    0.23
     society
    0.23
    éĨĴ
    0.23
    çļĦçĶŁåij½
    0.23
    Activations Density 0.910%

    No Known Activations