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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 42285
    Prev
    Next
    INDEX
    Explanations

    production

    np_max-act · gemini-2.0-flash

    The neuron activates on identifiers for runtime or build configurations—environment names like “production,” “development,” “test,” “Release,” etc.

    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
    Mus
    -0.07
    .school
    -0.07
    platz
    -0.06
    사를
    -0.06
    eriod
    -0.06
    rics
    -0.06
    .assertNot
    -0.06
    eg
    -0.06
    BeNull
    -0.06
     rescue
    -0.06
    POSITIVE LOGITS
     النو
    0.07
    .'),↵
    0.07
     Ole
    0.07
     sanitizer
    0.06
     Repo
    0.06
    UserName
    0.06
    Supply
    0.06
     trung
    0.06
    _histogram
    0.06
     Hopefully
    0.06
    Activations Density 0.006%

    No Known Activations