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. 8131
    Prev
    Next
    INDEX
    Explanations

    legal claims and court cases

    np_max-act · gemini-2.0-flash

    The neuron fires on formal legal‐document header tokens—court names, jurisdiction labels, docket numbers, filing dates, and related uppercase metadata.

    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
     nói
    -0.07
    값
    -0.07
    oyo
    -0.07
     tomatoes
    -0.06
    าเล
    -0.06
     trov
    -0.06
    _wall
    -0.06
    aju
    -0.06
     Dialogue
    -0.06
    YTE
    -0.06
    POSITIVE LOGITS
     irc
    0.06
     اروپا
    0.06
    SeekBar
    0.06
    _PO
    0.06
    (&$
    0.06
     Sn
    0.06
    _ability
    0.06
    Freq
    0.06
    cleanup
    0.06
     피
    0.06
    Activations Density 0.022%

    No Known Activations