© 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. 12659
    Prev
    Next
    INDEX
    Explanations

    tree

    np_max-act · gemini-2.0-flash

    The neuron is detecting mentions of data‐structure implementations—especially terms like “graph,” “tree,” “BinaryTree,” etc.—in code examples.

    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
    -delete
    -0.07
    ्‍
    -0.07
     emojis
    -0.06
     Над
    -0.06
    sass
    -0.06
    电视
    -0.06
     RelayCommand
    -0.06
     Shank
    -0.06
    setter
    -0.06
     Fletcher
    -0.06
    POSITIVE LOGITS
     Omni
    0.06
    umph
    0.06
    *dt
    0.06
     Functor
    0.06
    Geom
    0.06
    .Vector
    0.05
     Beverly
    0.05
     incompatible
    0.05
     εμφ
    0.05
     boosting
    0.05
    Activations Density 0.017%

    No Known Activations