© 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
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 128683
    Prev
    Next
    INDEX
    Explanations

    (

    np_max-act · gemini-2.0-flash

    This neuron fires on Angular-specific identifier strings—especially app/module names and state names (the quoted tokens in angular.module and $stateProvider.state calls).

    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
     subgroup
    -0.07
     mole
    -0.06
     inclusion
    -0.06
     στη
    -0.06
     amen
    -0.06
    CLUSION
    -0.06
    ISIBLE
    -0.06
     agency
    -0.06
    _POSITION
    -0.06
     ي
    -0.06
    POSITIVE LOGITS
     jednoho
    0.07
    asking
    0.07
     richness
    0.06
     licked
    0.06
    atat
    0.06
    844
    0.06
    914
    0.06
    jas
    0.06
    udev
    0.06
     Tory
    0.06
    Activations Density 0.002%

    No Known Activations