© 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
    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. 125950
    Prev
    Next
    INDEX
    Explanations

    Modality

    np_max-act · gemini-2.0-flash

    This neuron detects modal auxiliary verbs that express obligation or recommendation (e.g. “should,” “must,” 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
    くの
    -0.07
    elijk
    -0.07
    iou
    -0.06
     народ
    -0.06
    북도
    -0.06
    NullOrEmpty
    -0.06
    Mission
    -0.06
     Да
    -0.06
     شرق
    -0.06
     signifies
    -0.06
    POSITIVE LOGITS
     delaying
    0.07
     deposited
    0.06
    ạc
    0.06
     gamb
    0.06
    Owned
    0.06
    Delay
    0.06
    ída
    0.06
    _SETUP
    0.06
    QUI
    0.06
    CFG
    0.06
    Activations Density 0.016%

    No Known Activations