© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 57035
    Prev
    Next
    INDEX
    Explanations

    first

    np_max-act · gemini-2.0-flash

    This neuron detects ordinal indicators—especially the word “first”—marking initial occurrences in sequences (e.g. “first frame,” “first set,” 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
    _places
    -0.07
    	actor
    -0.07
    Italic
    -0.07
    elix
    -0.06
    CONTENT
    -0.06
     excer
    -0.06
     Malta
    -0.06
    jspx
    -0.06
     luc
    -0.06
    adir
    -0.06
    POSITIVE LOGITS
     projections
    0.07
    rw
    0.07
     projection
    0.06
    .....
    0.06
     already
    0.06
     narrative
    0.06
    ILING
    0.06
     λόγ
    0.06
    jian
    0.06
     пов
    0.06
    Activations Density 0.107%

    No Known Activations