© 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 EvalsExportsinterp-engineNEWAPI 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. 39097
    Prev
    Next
    INDEX
    Explanations

    singing together

    np_max-act · gemini-2.0-flash

    references to musical elements and themes of connection through music.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron responds to exclamatory cues inviting group participation—phrases like “Everybody together:” or calls to sing along that signal collective engagement.

    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
     liter
    -0.06
     tackles
    -0.06
    Cause
    -0.06
    reck
    -0.06
     ratio
    -0.06
     vowel
    -0.06
     operate
    -0.06
    STOP
    -0.06
    igm
    -0.06
     Holy
    -0.05
    POSITIVE LOGITS
    .blur
    0.08
     sortBy
    0.07
    melon
    0.07
    /Internal
    0.07
    浩
    0.07
    наслід
    0.07
     přístup
    0.07
    _combine
    0.06
     achievements
    0.06
     Başkanı
    0.06
    Activations Density 0.077%

    No Known Activations