© 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. 68868
    Prev
    Next
    INDEX
    Explanations

    )

    np_max-act · gemini-2.0-flash

    This neuron activates on the phrase “unstructured array,” i.e. it recognizes mentions of that specific data‐structure term.

    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
    Registrar
    -0.07
    чив
    -0.07
    _SERVER
    -0.07
    _stride
    -0.06
    聊
    -0.06
    VN
    -0.06
     slit
    -0.06
     Allocator
    -0.06
    Rotation
    -0.06
     Zur
    -0.06
    POSITIVE LOGITS
     green
    0.07
     motiv
    0.07
     clinging
    0.07
    Paris
    0.07
     eighteen
    0.06
     saldo
    0.06
    ідом
    0.06
     sapi
    0.06
     Muss
    0.06
    ant
    0.06
    Activations Density 0.101%

    No Known Activations