© 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. 107328
    Prev
    Next
    INDEX
    Explanations

    under

    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
    lei
    -0.29
     suites
    -0.27
    Fetcher
    -0.26
    éĺij
    -0.26
    spender
    -0.25
    lfw
    -0.25
    leep
    -0.25
    ikip
    -0.24
    meth
    -0.24
    lush
    -0.24
    POSITIVE LOGITS
    rica
    0.25
    ris
    0.24
    -Co
    0.24
    prior
    0.24
    线索
    0.24
     UB
    0.23
     Ult
    0.23
    chimp
    0.23
     اÙĦعÙĨ
    0.23
    ture
    0.23
    Activations Density 0.932%

    No Known Activations