© 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
    2. Qwen3-32B
    3. 32-RESID-BATCHTOPK-65K
    4. 53921
    Prev
    Next
    INDEX
    Explanations

    password is your responsibility. No Customer Service Representative knows or can retrieve your password for you. Representatives will never ask you for yourPairs.add(new BasicNameValuePair("login_password", "******")); /*'login_email' and 'login_password' i found html form fields name by viewing the page source of https://www.dropbox.com/login */ tryThe texts seem to involve forms, fields, passwords, login credentials, and potentially HTML elements. The `TOP_POSITIVE_LOGITS` list includes `.Scheme` and `digits` (and `_digits`), which could relate to specific data types or configurations often found in web forms or code.The `MAX_ACTIVATING_TOKENS` include `place`, `found`, `type`, `empty`, `open`, `all`, `one`, `voice` which could describe states or actions related to input fields or parameters. `TOKENS_AFTER_MAX_ACTIVATING_TOKEN` includes `html`, `automatically`, `request`, `channel`, which are common in technical contexts.login forms and credentials

    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
    ÑĢеÑģ
    -0.10
    åħ¨æĻ¯
    -0.09
    æłĩé¢ĺ
    -0.09
    plevel
    -0.09
    åı¯ä»¥éĢīæĭ©
    -0.08
    ImageRelation
    -0.08
    人çĶŁ
    -0.08
     Gros
    -0.08
    笾
    -0.08
    CloseOperation
    -0.08
    POSITIVE LOGITS
    çī¢è®°
    0.10
    BITS
    0.08
    çŁ¥æĻĵ
    0.08
    .Scheme
    0.08
    æĭ¿çĿĢ
    0.08
     digits
    0.08
    _digits
    0.08
    身份è¯ģ
    0.08
     Agencies
    0.08
    igos
    0.07
    Activations Density 0.326%

    No Known Activations