© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 7806
    Prev
    Next
    INDEX
    Explanations

    energy

    np_max-act · gemini-2.0-flash

    The neuron is triggered by occurrences of “energy” (especially when paired with “drink” or “drinks”), effectively detecting mentions of energy‐drink products.

    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
    iti
    -0.08
     Kut
    -0.07
    Mon
    -0.07
    _dead
    -0.07
     숙
    -0.07
    _XDECREF
    -0.07
    ційного
    -0.07
     Ferry
    -0.06
    Choice
    -0.06
    icipants
    -0.06
    POSITIVE LOGITS
     нату
    0.07
    	ev
    0.07
    PN
    0.06
    "$
    0.06
    _lv
    0.06
    _________________↵↵
    0.06
    oundation
    0.06
    (""));↵
    0.06
     ln
    0.06
    jack
    0.06
    Activations Density 0.012%

    No Known Activations