© 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. 129657
    Prev
    Next
    INDEX
    Explanations

    substrate direction

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of substrates (especially references to “substrate” or “substrate surfaces”) in technical patent text.

    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
     epilepsy
    -0.06
    ُن
    -0.06
    hind
    -0.06
    났다
    -0.06
    Controller
    -0.06
    ю
    -0.06
    ingham
    -0.06
    (dateTime
    -0.06
     cerebral
    -0.06
    fter
    -0.06
    POSITIVE LOGITS
    _unc
    0.06
     TRACE
    0.06
    历史
    0.06
    	resp
    0.06
    ρωπα
    0.06
    PRIMARY
    0.06
     Techn
    0.06
     演
    0.06
    (any
    0.06
     fibonacci
    0.05
    Activations Density 0.027%

    No Known Activations