Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Qwen2.5-7B-IT
    4. Resid Post - 131k
    5. 19-RESID-POST-AA
    6. 24688
    Prev
    Next
    INDEX
    Explanations

    :

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_19/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.19.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    河流域
    -0.07
     amateurs
    -0.07
     kształt
    -0.07
    鳘
    -0.07
    的情况下
    -0.07
    杵
    -0.07
     chooses
    -0.07
    ейств
    -0.07
    хват
    -0.07
    ⺟
    -0.07
    POSITIVE LOGITS
     ())↵
    0.08
     Compression
    0.07
    .ident
    0.07
     expressing
    0.07
    .createClass
    0.07
    _JUMP
    0.07
    ISED
    0.07
    _EXTENSION
    0.07
     федеральн
    0.07
    רב
    0.07
    Activations Density 0.001%

    No Known Activations