© 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. 22-TRANSCODER-HP
    7. 111587
    Prev
    Next
    INDEX
    Explanations

    .

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ATEST
    -0.28
    æĬķæĶ¾
    -0.27
     rid
    -0.25
    è¿ģ
    -0.25
    asive
    -0.25
    æľĢéļ¾
    -0.25
     polls
    -0.24
    lettes
    -0.24
    óÅĤ
    -0.24
    .endpoint
    -0.24
    POSITIVE LOGITS
    æĪijæĺ¯
    0.30
    orth
    0.29
    /am
    0.28
    uros
    0.28
     Moody
    0.27
    ParameterValue
    0.27
    åĽłä¸ºæĪij
    0.26
    å¦Ĥæľī
    0.25
    (calc
    0.25
     bail
    0.25
    Activations Density 0.003%

    No Known Activations