© 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. 85684
    Prev
    Next
    INDEX
    Explanations

    misuse

    np_max-act · gemini-2.0-flash

    The neuron activates on words indicating improper use, such as “abuse,” “misuse,” or similar terms.

    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
    /ca
    -0.07
    .Utils
    -0.06
    (instruction
    -0.06
    基
    -0.06
     Kou
    -0.06
     между
    -0.06
    вич
    -0.06
    unique
    -0.06
    elho
    -0.06
     tamanho
    -0.06
    POSITIVE LOGITS
     misuse
    0.08
     Domain
    0.07
     domain
    0.07
     Theft
    0.07
     bestowed
    0.07
     Stuttgart
    0.07
    享
    0.07
    BE
    0.06
    прав
    0.06
     intox
    0.06
    Activations Density 0.011%

    No Known Activations