© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 38010
    Prev
    Next
    INDEX
    Explanations

    know

    np_max-act · gemini-2.0-flash

    This neuron responds to first-person admission or opinion phrases—especially “I” followed by “do” or similar verbs expressing uncertainty or personal judgment (e.g. “I do not know,” “I think”).

    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
     hash
    -0.07
     "%
    -0.07
    omics
    -0.07
    _el
    -0.07
    _VERBOSE
    -0.07
     ห
    -0.06
     isNaN
    -0.06
    This
    -0.06
    goog
    -0.06
    _counter
    -0.06
    POSITIVE LOGITS
     publishes
    0.07
     stol
    0.06
     regularization
    0.06
    _attention
    0.06
    lásil
    0.06
    ');?>↵
    0.06
     νό
    0.06
     recommended
    0.06
    (curl
    0.06
    onet
    0.06
    Activations Density 0.076%

    No Known Activations