© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 9224
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of legal “possession of” charges (e.g. possession of a weapon or controlled substance).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Cz
    -0.07
    атки
    -0.07
    interopRequire
    -0.06
    _into
    -0.06
    олит
    -0.06
     ascent
    -0.06
    ının
    -0.06
     الوط
    -0.06
    >>>>>>>
    -0.06
     Ricky
    -0.06
    POSITIVE LOGITS
    rades
    0.07
    ��
    0.07
    	flash
    0.06
     Wanted
    0.06
    claimer
    0.06
    izontal
    0.06
    proxy
    0.06
    电子
    0.06
    分别
    0.06
    .system
    0.06
    Activations Density 0.004%

    No Known Activations