© 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. 11-RESID-POST-AA
    11. 109887
    Prev
    Next
    INDEX
    Explanations

    primary

    np_max-act · gemini-2.0-flash

    This neuron activates on CSS/SCSS color definitions and references—that is, tokens involved in setting or naming colors (e.g. color variables, color-related property names, and gradient definitions).

    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
     luôn
    -0.07
    ером
    -0.07
    Stored
    -0.07
     enlightened
    -0.07
    教育
    -0.07
    °C
    -0.06
    osyal
    -0.06
    /count
    -0.06
     Milton
    -0.06
     homogeneous
    -0.06
    POSITIVE LOGITS
     По
    0.06
     ПО
    0.06
     Guerr
    0.06
     Orange
    0.06
    etz
    0.06
    воз
    0.06
     cand
    0.06
    .'</
    0.05
     regeneration
    0.05
    /conf
    0.05
    Activations Density 0.006%

    No Known Activations