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. 43761
    Prev
    Next
    INDEX
    Explanations

    code, programming, technical documentation

    np_max-act · gemini-2.0-flash

    instances of fixed-size arrays and constants in code.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron fires on short all-caps tokens—especially Roman‐numeral style tokens like “II.”

    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
     elder
    -0.06
     tự
    -0.06
     थ
    -0.06
    _Native
    -0.06
    الث
    -0.06
    farm
    -0.06
    čně
    -0.06
    -Pack
    -0.06
     数
    -0.06
     Bay
    -0.06
    POSITIVE LOGITS
     waveform
    0.08
     Installer
    0.07
    yles
    0.07
    hdl
    0.06
     свій
    0.06
     Magnum
    0.06
    .Dep
    0.06
     XHTML
    0.06
    BMW
    0.06
    INO
    0.06
    Activations Density 0.000%

    No Known Activations