Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 41800
    Prev
    Next
    INDEX
    Explanations

    medical patients

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of “patient,” especially within phrases like “any patient.”

    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
     outFile
    -0.07
     dp
    -0.06
     yaşam
    -0.06
    бас
    -0.06
     Dat
    -0.06
    avicon
    -0.06
    facet
    -0.06
     taşım
    -0.06
    _TABLE
    -0.06
     Elis
    -0.06
    POSITIVE LOGITS
    _Channel
    0.06
     indebted
    0.06
     urb
    0.06
    ฤษ
    0.06
     richness
    0.06
     currentIndex
    0.06
    unsqueeze
    0.06
    (mean
    0.06
    Persist
    0.06
    集团
    0.06
    Activations Density 0.174%

    No Known Activations