© 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. 55934
    Prev
    Next
    INDEX
    Explanations

    negative sentiments

    np_max-act · gemini-2.0-flash

    The neuron responds to technical academic research jargon (e.g., terms like “provable,” “security,” “proofs,” “research,” “technical”).

    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
    ffer
    -0.07
     formato
    -0.07
    ANDOM
    -0.06
    _prop
    -0.06
     rozh
    -0.06
    inp
    -0.06
    cone
    -0.06
     curated
    -0.06
     Fruit
    -0.06
    cedure
    -0.06
    POSITIVE LOGITS
     CONVERT
    0.06
    ителем
    0.06
    Deserialize
    0.06
    engine
    0.06
     [];
↵
    0.06
    _DEFAULT
    0.06
     Dup
    0.06
     eggs
    0.06
    .deb
    0.06
     늘
    0.06
    Activations Density 1.334%

    No Known Activations