Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 519
    Prev
    Next
    INDEX
    Explanations

    time ranges in months

    np_max-act · gemini-2.0-flash

    This neuron appears never to activate on any tokens—it doesn’t detect any meaningful pattern (i.e. it’s effectively “dead”).

    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
     reef
    -0.07
    UsingEncoding
    -0.06
    “When
    -0.06
    groupon
    -0.06
    <void
    -0.06
    anel
    -0.06
    _LINK
    -0.06
     stamp
    -0.06
     sanction
    -0.06
    -react
    -0.06
    POSITIVE LOGITS
    ブラ
    0.07
    Intern
    0.06
     Ins
    0.06
    Brown
    0.06
    итай
    0.06
     بودند
    0.06
    phins
    0.06
    indent
    0.06
    .userid
    0.06
     Calendar
    0.06
    Activations Density 0.006%

    No Known Activations