© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 38149
    Prev
    Next
    INDEX
    Explanations

    bridge and bridging

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ally
    -0.11
    tract
    -0.11
    ë£Į
    -0.11
    ities
    -0.11
    åĭĥåĭĥ
    -0.10
    ãģĦãģĨ
    -0.10
    .Encoding
    -0.09
    istically
    -0.09
    igue
    -0.09
    uv
    -0.09
    POSITIVE LOGITS
    ï¸
    0.11
    åłį
    0.10
    hands
    0.09
    ues
    0.09
    -mounted
    0.09
    墩
    0.09
    inar
    0.09
    Ñģ
    0.09
    iens
    0.09
    edList
    0.09
    Activations Density 0.029%

    No Known Activations