© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 73612
    Prev
    Next
    INDEX
    Explanations

    global

    np_max-act · gemini-2.0-flash

    This neuron fires on C# auto‐generated resource accessor code, in particular the declarations of the “Resources” class and its ResourceManager (e.g. the strongly-typed resource class in a .Properties namespace).

    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
     Yo
    -0.07
     smoked
    -0.06
     Dodgers
    -0.06
     onLoad
    -0.06
     iniciar
    -0.06
    /static
    -0.06
    ned
    -0.06
     nhiệt
    -0.06
    -bin
    -0.06
    armor
    -0.06
    POSITIVE LOGITS
     inspector
    0.06
     Network
    0.06
    _SCENE
    0.06
    的一
    0.06
     склада
    0.06
     суспіль
    0.06
    acic
    0.06
    utable
    0.06
     uLocal
    0.06
    sey
    0.06
    Activations Density 0.004%

    No Known Activations