© 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. 129714
    Prev
    Next
    INDEX
    Explanations

    withdrawal

    np_max-act · gemini-2.0-flash

    The neuron chiefly detects mentions of withdrawal or detoxification (e.g., “withdrawal,” “syndrome,” “detox,” “delirium”).

    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
     Aless
    -0.08
     먼저
    -0.07
    .policy
    -0.06
     }),↵
    -0.06
    (DATA
    -0.06
     Hồng
    -0.06
     Al
    -0.06
    _detector
    -0.06
     dataList
    -0.06
    Areas
    -0.06
    POSITIVE LOGITS
    .Fore
    0.07
    ****************************************************************************
    0.07
     imposing
    0.07
    hasOne
    0.06
    _interface
    0.06
     boredom
    0.06
     stakes
    0.06
    phies
    0.06
     niż
    0.06
     طراحی
    0.06
    Activations Density 0.002%

    No Known Activations