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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 54991
    Prev
    Next
    INDEX
    Explanations

    or

    np_max-act · gemini-2.0-flash

    The neuron fires on terms referring to legal procedural actions or motions.

    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
    xfb
    -0.07
     Hans
    -0.06
    Œ
    -0.06
    RTOS
    -0.06
    usal
    -0.06
    剑
    -0.06
    _res
    -0.06
    anı
    -0.06
    clubs
    -0.06
    �
    -0.06
    POSITIVE LOGITS
     or
    0.07
    <k
    0.06
    ;/*
    0.06
     حالة
    0.06
     Sofa
    0.06
    Ranked
    0.06
    -alt
    0.06
     Por
    0.06
    highest
    0.06
     genetically
    0.06
    Activations Density 0.011%

    No Known Activations