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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 84469
    Prev
    Next
    INDEX
    Explanations

    at

    np_max-act · gemini-2.0-flash

    The neuron fires on geographic place names, especially locations introduced by prepositions like “at” or “in” in event contexts (e.g. places of birth or death).

    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
    .False
    -0.08
     gg
    -0.07
     rời
    -0.07
     Sé
    -0.07
     Belmont
    -0.07
    ois
    -0.06
     rè
    -0.06
     خوان
    -0.06
     Row
    -0.06
    amples
    -0.06
    POSITIVE LOGITS
    fiction
    0.07
     modify
    0.07
     oppressed
    0.07
    MES
    0.06
    ILLED
    0.06
    ';↵↵↵↵
    0.06
    ılacak
    0.06
     sulla
    0.06
     comunidad
    0.06
    stop
    0.06
    Activations Density 0.007%

    No Known Activations