© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 73779
    Prev
    Next
    INDEX
    Explanations

    algorithm

    np_max-act · gemini-2.0-flash

    The neuron activates on technical terms referring to algorithmic or formulaic procedures (e.g. “algorithm,” “formula,” “secret formula”).

    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
    .*?)
    -0.07
    ainted
    -0.07
     ders
    -0.07
     Alarm
    -0.07
     grac
    -0.06
     Cres
    -0.06
    cit
    -0.06
     початку
    -0.06
    emiz
    -0.06
    153
    -0.06
    POSITIVE LOGITS
     REGION
    0.07
    ουν
    0.07
     grains
    0.06
     quotas
    0.06
     흐
    0.06
     somew
    0.06
    .")↵
    0.06
    ierarchy
    0.06
    -aff
    0.06
     prueba
    0.06
    Activations Density 0.013%

    No Known Activations