© 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. 22047
    Prev
    Next
    INDEX
    Explanations

    experimental stages and setups

    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
    -ignore
    -0.09
    巨人
    -0.08
    .Typed
    -0.08
    è§Ĩè§ī
    -0.08
    owo
    -0.08
     LOD
    -0.08
    åĽ½åºĨ
    -0.08
    çķĻ
    -0.08
    ä¸Ģ线
    -0.08
    触
    -0.08
    POSITIVE LOGITS
    è¯ķéªĮ
    0.18
    å®ŀéªĮ
    0.16
    試
    0.14
    å®ŀéªĮ室
    0.13
    æµĭè¯ķ
    0.13
    åİŁåŀĭ
    0.12
    è¯ķéªĮåĮº
    0.12
     experiment
    0.12
    )test
    0.12
    experimental
    0.11
    Activations Density 0.201%

    No Known Activations