© 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. 129665
    Prev
    Next
    INDEX
    Explanations

    a

    np_max-act · gemini-2.0-flash

    This neuron detects the structural markers of step‐by‐step instructions (e.g. “Step,” step numbers, and the accompanying colons).

    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
     developers
    -0.07
     cone
    -0.06
     hồ
    -0.06
    (E
    -0.06
    dlg
    -0.06
    意见
    -0.06
     >::
    -0.06
     pope
    -0.06
    DEN
    -0.06
     initiated
    -0.06
    POSITIVE LOGITS
     Шев
    0.06
     отп
    0.06
    .Ignore
    0.06
     �
    0.06
     première
    0.06
    larıyla
    0.06
     đăng
    0.06
     všichni
    0.06
    _nth
    0.06
    sgiving
    0.06
    Activations Density 0.013%

    No Known Activations