© 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. 53544
    Prev
    Next
    INDEX
    Explanations

    foreign words

    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
    zeichnet
    -0.28
    .validator
    -0.27
    ilarity
    -0.27
    æı´
    -0.25
     plaque
    -0.25
    å¯ĩ
    -0.25
    ette
    -0.24
    æ®Ľ
    -0.24
    appa
    -0.23
    zeich
    -0.23
    POSITIVE LOGITS
    hr
    0.35
    /hr
    0.28
    	hr
    0.28
    esModule
    0.27
    hra
    0.27
     Lig
    0.26
     hex
    0.26
    HR
    0.26
    uan
    0.26
    hex
    0.26
    Activations Density 0.034%

    No Known Activations