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. 8702
    Prev
    Next
    INDEX
    Explanations

    authority

    np_max-act · gemini-2.0-flash

    The neuron fires on the formal “convening authority” approval language in military judicial opinions (e.g. mentions of the convening authority’s decision to approve or adjudge the sentence).

    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
    ниця
    -0.07
     ("<
    -0.06
     HEL
    -0.06
     inde
    -0.06
    (CON
    -0.06
     filmed
    -0.06
    _car
    -0.06
     пози
    -0.06
    IRCLE
    -0.06
     ống
    -0.06
    POSITIVE LOGITS
    history
    0.07
     unwind
    0.06
     μπορεί
    0.06
    ซ
    0.06
    ев
    0.06
    κολ
    0.06
    िव
    0.06
    icularly
    0.06
    ян
    0.06
    (mx
    0.06
    Activations Density 0.001%

    No Known Activations