© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 42713
    Prev
    Next
    INDEX
    Explanations

    core

    np_max-act · gemini-2.0-flash

    This neuron activates on hardware spec tokens—numbers and their related units or terms (e.g. CPU, cores, GB, GHz)—essentially spotting mentions of technical system requirements.

    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
    reiben
    -0.07
    stm
    -0.06
    ापन
    -0.06
    slot
    -0.06
     Boston
    -0.06
    ША
    -0.06
    ║
    -0.06
     unreal
    -0.06
    пат
    -0.06
     fatal
    -0.06
    POSITIVE LOGITS
    cmpeq
    0.07
     отказ
    0.07
     noexcept
    0.07
     ==============================================================
    0.06
    抓
    0.06
    .Animation
    0.06
    Mac
    0.06
    еры
    0.06
    ана
    0.06
    �
    0.06
    Activations Density 0.005%

    No Known Activations