© 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 EvalsExportsinterp-engineNEWAPI 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. 57892
    Prev
    Next
    INDEX
    Explanations

    created

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens in code comment headers (e.g. version numbers, dates, or timestamps in Javadoc‐style comments).

    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
    /gif
    -0.07
    .Tree
    -0.07
    etadata
    -0.07
    ัฒ
    -0.07
    Into
    -0.07
    .")
↵
    -0.07
    Types
    -0.06
     USHORT
    -0.06
     equivalent
    -0.06
    ५
    -0.06
    POSITIVE LOGITS
     knex
    0.07
     fetchData
    0.06
    irma
    0.06
     перер
    0.06
    preneur
    0.06
     Öz
    0.06
    ційний
    0.06
    ITUDE
    0.06
    .findIndex
    0.06
     тек
    0.06
    Activations Density 0.003%

    No Known Activations