© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 5922
    Prev
    Next
    INDEX
    Explanations

    determine

    np_max-act · gemini-2.0-flash

    This neuron activates on the main directive verb (“determine”) at the start of the instruction.

    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
     "+"
    -0.06
    ollections
    -0.06
    อนไลน
    -0.06
    OPSIS
    -0.06
    /how
    -0.06
     لع
    -0.06
    Girls
    -0.06
    Templates
    -0.06
    _oauth
    -0.06
     onSubmit
    -0.06
    POSITIVE LOGITS
    ufac
    0.06
     configurable
    0.06
     instrumentation
    0.06
     rabbit
    0.06
     distinguishing
    0.06
     counseling
    0.06
    _delta
    0.06
     definitive
    0.06
     Zimmerman
    0.06
     malaria
    0.06
    Activations Density 0.006%

    No Known Activations