© 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. 89909
    Prev
    Next
    INDEX
    Explanations

    {

    np_max-act · gemini-2.0-flash

    The neuron activates on the declaration of functions or methods (e.g. the “public … (…)” and “run()” signatures).

    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
    _documents
    -0.06
     ż
    -0.06
    Coverage
    -0.06
    _SPEED
    -0.06
    NX
    -0.06
    igital
    -0.06
    _h
    -0.06
     Moment
    -0.06
     canoe
    -0.06
     OBJECT
    -0.06
    POSITIVE LOGITS
     کسانی
    0.07
     ragazza
    0.07
     immense
    0.07
    .Undef
    0.07
     आर
    0.07
     Τζ
    0.06
     afforded
    0.06
    orderId
    0.06
    _province
    0.06
     Пари
    0.06
    Activations Density 0.030%

    No Known Activations