© 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. 126076
    Prev
    Next
    INDEX
    Explanations

    composite

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on mentions of “composite” (and closely related terms like “composite materials,” “composites,” and “fiber-reinforced composite”) in technical contexts.

    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
    μό
    -0.06
    监督
    -0.06
    _a
    -0.06
    .fil
    -0.06
    datal
    -0.06
    _BOLD
    -0.06
     broadband
    -0.06
     уб
    -0.06
     decreases
    -0.06
    svc
    -0.06
    POSITIVE LOGITS
    Guy
    0.07
     tentative
    0.07
    usahaan
    0.07
     improper
    0.07
     Hawaiian
    0.07
     trung
    0.07
    _inventory
    0.07
    chied
    0.07
    ―
    0.06
    Researchers
    0.06
    Activations Density 0.008%

    No Known Activations