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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 13741
    Prev
    Next
    INDEX
    Explanations

    Positive sentiment

    np_max-act · gemini-2.0-flash

    This neuron activates on descriptive adjectives (especially positive or evaluative modifiers) that highlight qualities or scope (e.g. “diverse,” “unique,” “thriving,” “engaging,” “vast”).

    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
    CppGeneric
    -0.06
     most
    -0.06
     είναι
    -0.06
     Laguna
    -0.06
     состоянии
    -0.06
     erotici
    -0.06
    眠
    -0.06
     Veteran
    -0.06
     Elias
    -0.06
     meer
    -0.06
    POSITIVE LOGITS
    ighbor
    0.06
     checkpoints
    0.06
    .createUser
    0.06
    .getID
    0.06
     liner
    0.06
     researcher
    0.06
    	token
    0.06
     cliffs
    0.06
    ۱۹۸
    0.06
    constraints
    0.06
    Activations Density 0.130%

    No Known Activations