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. 42251
    Prev
    Next
    INDEX
    Explanations

    Code snippets

    np_max-act · gemini-2.0-flash

    This neuron detects BI query language keywords and constructs (e.g. SET, CREATE, MEMBER, CALCULATE, SAMEPERIODLASTYEAR, filters, date/table references) in MDX/DAX code.

    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
    Triangle
    -0.07
     ци
    -0.07
    .Touch
    -0.07
    Newton
    -0.06
    pec
    -0.06
    altar
    -0.06
     плот
    -0.06
     då
    -0.06
    Grow
    -0.06
     Dx
    -0.06
    POSITIVE LOGITS
     enfrent
    0.07
     intriguing
    0.06
     náp
    0.06
    ?]
    0.06
     realloc
    0.06
     اجتماع
    0.06
    	number
    0.06
     prime
    0.06
     οι
    0.06
    (TokenType
    0.06
    Activations Density 0.012%

    No Known Activations