© 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 EvalsExportsAPI 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. 38045
    Prev
    Next
    INDEX
    Explanations

    conversations

    np_max-act · gemini-2.0-flash

    The neuron fires on the assistant’s offers of help or assistance (e.g. “assist,” “help” in “How can I assist you today?”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    instances where the assistant offers help or asks how it can assist (phrases like "How can I assist/help you today?").

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
     lifting
    -0.07
     GetUser
    -0.06
    ח
    -0.06
     Gel
    -0.06
    HU
    -0.06
    -0.06
     gel
    -0.06
    lr
    -0.06
     thermometer
    -0.06
    Side
    -0.06
    POSITIVE LOGITS
     typography
    0.06
    //
    0.06
     Emil
    0.06
    	volatile
    0.06
     boots
    0.06
    ($__
    0.06
    -root
    0.06
    .root
    0.06
    	glUniform
    0.06
     uso
    0.06
    Activations Density 0.111%

    No Known Activations