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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 54026
    Prev
    Next
    INDEX
    Explanations

    memory allocation

    np_max-act · gemini-2.0-flash

    The neuron is detecting mentions of pointers and dynamic memory management (e.g. '*' declarations, malloc/free, new/delete).

    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
     смерти
    -0.07
     капіт
    -0.06
     harmful
    -0.06
    สถาน
    -0.06
    堆
    -0.06
    _pv
    -0.06
    elerinde
    -0.06
     laboratory
    -0.06
     wages
    -0.06
     playa
    -0.06
    POSITIVE LOGITS
     vin
    0.07
    ":""
    0.06
    /ss
    0.06
    StringUtils
    0.06
     creo
    0.06
     chu
    0.06
    ่ก
    0.06
    ={
    0.06
    ΙΑΚ
    0.06
    ')}>↵
    0.06
    Activations Density 0.009%

    No Known Activations