© 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. 12999
    Prev
    Next
    INDEX
    Explanations

    morph

    np_max-act · gemini-2.0-flash

    The neuron detects morphology‐related terms (words containing the root “morph,” e.g. morphological, morphogenesis).

    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
    تص
    -0.07
     घर
    -0.06
    WAIT
    -0.06
    _summary
    -0.06
    ABC
    -0.06
     Attacks
    -0.06
     된
    -0.06
     penalty
    -0.06
    еру
    -0.06
     группы
    -0.06
    POSITIVE LOGITS
    olygon
    0.06
    ائر
    0.06
     psychologically
    0.06
    www
    0.06
    バ
    0.06
     percept
    0.06
    ................................................................
    0.06
     sag
    0.06
    select
    0.06
    vements
    0.06
    Activations Density 0.047%

    No Known Activations