© 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. 110144
    Prev
    Next
    INDEX
    Explanations

    sure

    np_max-act · gemini-2.0-flash

    The neuron activates on precautionary or warning phrases—especially “make sure,” “be sure,” and similar calls to back up data or take safety steps.

    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
    Assert
    -0.07
    _aligned
    -0.07
    決定
    -0.07
     DLL
    -0.06
    LP
    -0.06
    ilerden
    -0.06
    ó
    -0.06
    uro
    -0.06
     steel
    -0.06
    ILD
    -0.06
    POSITIVE LOGITS
     cigaret
    0.08
    NASDAQ
    0.06
     GOD
    0.06
     groot
    0.06
    .GetOrdinal
    0.06
     seedu
    0.06
    Donate
    0.06
     dislike
    0.06
    itta
    0.06
     WebGL
    0.06
    Activations Density 0.061%

    No Known Activations