© 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. 89811
    Prev
    Next
    INDEX
    Explanations

    relationships and stories

    np_max-act · gemini-2.0-flash

    The neuron detects content-word tokens within the user’s enumerated “themes of books I typically read” list (e.g. sword, magic, dragon, space, betrayal, etc.).

    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
     settle
    -0.07
    pak
    -0.07
     qr
    -0.07
     doğrudan
    -0.07
    .drop
    -0.07
    эт
    -0.07
     plagiarism
    -0.06
     použí
    -0.06
    045
    -0.06
     bounds
    -0.06
    POSITIVE LOGITS
    MERCHANTABILITY
    0.07
    ambient
    0.06
     сопров
    0.06
    (gl
    0.06
    isyon
    0.06
     stretches
    0.06
    体育
    0.06
    Intern
    0.06
     Ekon
    0.06
    .addEdge
    0.06
    Activations Density 0.040%

    No Known Activations