© 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. 658
    Prev
    Next
    INDEX
    Explanations

    commonly

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of routine or typical methods by firing on the phrase “commonly used” (as in “commonly used method/route”).

    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
     sculpt
    -0.07
    ceptor
    -0.06
    _Trans
    -0.06
     بسیار
    -0.06
    Rio
    -0.06
    ::
    -0.06
     Bec
    -0.06
    .IN
    -0.06
     ประเทศ
    -0.06
    .vertex
    -0.06
    POSITIVE LOGITS
     mA
    0.07
     frequent
    0.07
    呼
    0.06
     favourite
    0.06
     baise
    0.06
     accordance
    0.06
    手
    0.06
    thest
    0.06
     contacting
    0.06
    enido
    0.06
    Activations Density 0.020%

    No Known Activations