© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 91484
    Prev
    Next
    INDEX
    Explanations

    OR

    np_max-act · gemini-2.0-flash

    The neuron activates on words in software license disclaimers that list types of liability or damages (e.g. “OR,” “DATA,” “PROFITS,” “BUSINESS INTERRUPTION”), effectively spotting the enumeration of permitted/excluded damages.

    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
    атов
    -0.07
    ̉
    -0.07
     yen
    -0.07
    senal
    -0.07
    umno
    -0.06
    localctx
    -0.06
    Conf
    -0.06
     müm
    -0.06
    Injector
    -0.06
     Yen
    -0.06
    POSITIVE LOGITS
     huz
    0.06
     benchmarks
    0.06
    ีเม
    0.06
    _articles
    0.06
    _MESSAGE
    0.06
     Transparent
    0.06
     Би
    0.06
     χα
    0.06
    _NAV
    0.06
     Tooltip
    0.06
    Activations Density 0.002%

    No Known Activations