© 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. 13425
    Prev
    Next
    INDEX
    Explanations

    fibonacci

    np_max-act · gemini-2.0-flash

    The neuron activates on code-related mentions of “fibonacci” (and related terms like “calculating,” “nth,” etc.), i.e. when the text refers to computing the Fibonacci sequence in 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
    ція
    -0.07
    -pencil
    -0.06
    .ast
    -0.06
     Tristan
    -0.06
     Millenn
    -0.06
     USER
    -0.06
     NONE
    -0.06
     guru
    -0.06
     inputFile
    -0.06
     UserRole
    -0.06
    POSITIVE LOGITS
     Governors
    0.07
    terra
    0.07
    (),
↵
    0.06
    ardless
    0.06
     thrilled
    0.06
     fascinating
    0.06
     collections
    0.06
    ↵            ↵
    0.06
    věd
    0.06
     engineered
    0.06
    Activations Density 0.013%

    No Known Activations