© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 92114
    Prev
    Next
    INDEX
    Explanations

    trees

    np_max-act · gemini-2.0-flash

    This neuron activates on descriptive, size-oriented adjectives paired with natural elements—especially words like “tall” or “slender” modifying plants or trees.

    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
    _countries
    -0.07
    cheap
    -0.07
    рами
    -0.07
     NFL
    -0.06
     GetMessage
    -0.06
     kidd
    -0.06
     preseason
    -0.06
     prosecuting
    -0.06
    _sc
    -0.06
    (Yii
    -0.06
    POSITIVE LOGITS
     tree
    0.08
     trees
    0.08
     Tree
    0.07
     Trees
    0.07
     Routes
    0.06
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    0.06
     iso
    0.06
     TREE
    0.06
    ойчив
    0.06
    UnitTest
    0.06
    Activations Density 0.013%

    No Known Activations