© 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. 88997
    Prev
    Next
    INDEX
    Explanations

    top

    np_max-act · gemini-2.0-flash

    The neuron fires on mentions of a component’s spatial orientation—particularly references to the “top surface” (and similar top/bottom surface descriptors).

    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
     squads
    -0.07
     должна
    -0.07
    against
    -0.06
     бит
    -0.06
     Φ
    -0.06
    CLUDED
    -0.06
    controller
    -0.06
     blockIdx
    -0.06
     Gir
    -0.06
    igDecimal
    -0.06
    POSITIVE LOGITS
     aliqu
    0.06
     Speaker
    0.06
     og
    0.06
     retrie
    0.06
    卓
    0.06
    �
    0.06
     UIControl
    0.06
    idepress
    0.06
    enedor
    0.06
    uil
    0.06
    Activations Density 0.009%

    No Known Activations