© 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. 55587
    Prev
    Next
    INDEX
    Explanations

    x or a

    np_max-act · gemini-2.0-flash

    The neuron is detecting mathematical expressions, in particular numeric tokens, variable names, and the equals sign that form simple equations.

    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
     OP
    -0.06
     isConnected
    -0.06
    getItem
    -0.06
     vyt
    -0.06
    فل
    -0.06
    利用
    -0.06
     analyzing
    -0.06
    model
    -0.06
    平
    -0.06
     Hoy
    -0.06
    POSITIVE LOGITS
     aesthetics
    0.08
    ์ม
    0.06
    -б
    0.06
     ----------------------------------------------------------------------↵
    0.06
    γεν
    0.06
     slots
    0.06
     ainsi
    0.06
     rowIndex
    0.06
     тен
    0.06
    至
    0.06
    Activations Density 0.003%

    No Known Activations