© 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. 6590
    Prev
    Next
    INDEX
    Explanations

    practice

    np_max-act · gemini-2.0-flash

    The neuron activates on abstract, philosophical keywords that discuss fundamental nature or essence (e.g. “fundamental,” “nature,” “reality,” “illusory,” “perspective,” “entity,” “achieved”).

    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
    ERY
    -0.07
    .__
    -0.06
    (array
    -0.06
     tmp
    -0.06
    东西
    -0.06
     expr
    -0.06
    talya
    -0.06
    aker
    -0.06
    リー
    -0.06
    หม
    -0.06
    POSITIVE LOGITS
     क
    0.07
    _RAD
    0.07
    云
    0.07
     inception
    0.07
    .Reset
    0.06
    :async
    0.06
    (Resources
    0.06
    ;">
    ↵
    0.06
    ADDE
    0.06
    .TextAlignment
    0.06
    Activations Density 0.212%

    No Known Activations