© 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. 33-TRANSCODER-HP
    7. 53109
    Prev
    Next
    INDEX
    Explanations

    ÑĥÑİÑī

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    è°Ī
    -0.29
    åĸĥ
    -0.28
     Reno
    -0.28
    æīĵåĵį
    -0.26
     der
    -0.26
    uations
    -0.25
    getto
    -0.25
    ãģªãģı
    -0.24
    AtPath
    -0.24
    ÙĦÙĪ
    -0.24
    POSITIVE LOGITS
     das
    0.59
    das
    0.49
    Das
    0.44
     die
    0.42
     Das
    0.42
    	die
    0.37
     ÑįÑĤоÑĤ
    0.35
     ÑįÑĤÑĥ
    0.35
    die
    0.35
    æīĭä¸ŃçļĦ
    0.35
    Activations Density 0.033%

    No Known Activations