© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 15-RESID-POST-AA
    11. 89231
    Prev
    Next
    INDEX
    Explanations

    technical discussions

    np_max-act · gemini-2.0-flash

    The neuron detects procedural instructional language—imperative/how-to steps and command-oriented explanations (e.g., "To …", commands, and step-by-step directions).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     :↵
    -0.07
     elites
    -0.07
     ::↵
    -0.06
    Ấ
    -0.06
     throat
    -0.06
     carro
    -0.06
    "]:↵
    -0.06
    .'),↵
    -0.06
     oto
    -0.06
    %.↵
    -0.06
    POSITIVE LOGITS
    дан
    0.06
    appeared
    0.06
    われた
    0.06
     competitions
    0.06
     дан
    0.06
    _checkpoint
    0.06
     unbiased
    0.06
    _message
    0.06
    furt
    0.06
     lasc
    0.06
    Activations Density 0.362%

    No Known Activations