Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 44125
    Prev
    Next
    INDEX
    Explanations

    Avoiding rapid, repeated actions

    np_max-act · gemini-2.0-flash

    The neuron detects language around rapid or “too‐fast” user actions—especially clicking or login/post attempts that must be rate‐limited.

    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
     Wand
    -0.06
     accessor
    -0.06
     f
    -0.06
    Ver
    -0.06
     захист
    -0.06
    지원
    -0.06
    potential
    -0.06
    663
    -0.06
    getOption
    -0.06
     webinar
    -0.06
    POSITIVE LOGITS
    asyarakat
    0.07
     gdy
    0.07
     bach
    0.06
     tarihi
    0.06
    itchen
    0.06
    medical
    0.06
     ForCanBeConverted
    0.06
     assures
    0.06
     ocor
    0.06
    JEXEC
    0.06
    Activations Density 0.018%

    No Known Activations