© 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. 42364
    Prev
    Next
    INDEX
    Explanations

    it

    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
     Deal
    -0.31
    è¦ģèĢĥèĻij
    -0.30
    触åıĬ
    -0.30
     touches
    -0.29
    lie
    -0.29
    éģĩä¸Ĭ
    -0.29
    æݥ触åĪ°
    -0.29
     crosses
    -0.29
     exposures
    -0.28
    触æij¸
    -0.28
    POSITIVE LOGITS
     all
    0.40
    æīĢæľī
    0.38
    _all
    0.35
     All
    0.31
    	all
    0.30
    All
    0.29
    /all
    0.29
     вÑģе
    0.29
    æīĢæľīçļĦ
    0.28
    _ALL
    0.27
    Activations Density 0.008%

    No Known Activations