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. 125758
    Prev
    Next
    INDEX
    Explanations

    in

    np_max-act · gemini-2.0-flash

    references to notable individuals, including politicians and athletes.

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

    This neuron is highly activated by numeric expressions—years, ranks, statistics, and other digit‐based tokens.

    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
    ustomed
    -0.06
    ازات
    -0.06
    otle
    -0.06
    -messages
    -0.06
    」と
    -0.06
     stale
    -0.06
     グ
    -0.06
     Slots
    -0.06
    gabe
    -0.06
    teborg
    -0.06
    POSITIVE LOGITS
    崎
    0.07
    acts
    0.06
    чної
    0.06
     devast
    0.06
    chester
    0.06
     وزار
    0.06
     *>(
    0.06
    archy
    0.06
     náz
    0.06
    เทศ
    0.06
    Activations Density 0.099%

    No Known Activations