© 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
    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. 109433
    Prev
    Next
    INDEX
    Explanations

    this

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on deictic/context‐setting terms—words like “this,” “our,” and especially “world” (and their equivalents in other languages).

    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
    éné
    -0.08
    ovací
    -0.07
    _element
    -0.06
     refriger
    -0.06
    iče
    -0.06
    -in
    -0.06
     \\
    -0.06
    ajo
    -0.06
    -elements
    -0.06
    in
    -0.06
    POSITIVE LOGITS
     cripp
    0.07
    .headers
    0.06
     this
    0.06
    _offsets
    0.06
    .Session
    0.06
     대해
    0.06
     ways
    0.06
     Didn
    0.06
    更多
    0.06
    Để
    0.06
    Activations Density 0.030%

    No Known Activations