© 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. 108496
    Prev
    Next
    INDEX
    Explanations

    Clauses starting with "that/which/whose"

    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
    ä¸ŃçļĦ
    -0.27
    ")).
    -0.26
    )").
    -0.24
    azu
    -0.24
    ')).
    -0.24
    duino
    -0.24
    strup
    -0.24
    å·²ä¹ħ
    -0.23
    ä¸ŃåİŁ
    -0.23
    laces
    -0.23
    POSITIVE LOGITS
     that
    0.84
    that
    0.71
     whose
    0.68
    	that
    0.60
    _that
    0.58
     which
    0.56
    whose
    0.56
    _THAT
    0.55
     That
    0.53
    which
    0.51
    Activations Density 1.275%

    No Known Activations