© 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
    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. 74741
    Prev
    Next
    INDEX
    Explanations

    comparative, ending in "er"

    np_max-act · gemini-2.0-flash

    The neuron activates on comparative adjectives (words in the "-er" form).

    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
    藝
    -0.07
    will
    -0.07
     obtained
    -0.07
    v
    -0.07
    bout
    -0.07
     onView
    -0.07
    would
    -0.06
    ับ
    -0.06
    ill
    -0.06
    ould
    -0.06
    POSITIVE LOGITS
     stronger
    0.12
     wider
    0.12
    Higher
    0.12
     larger
    0.12
     faster
    0.12
    er
    0.12
    ER
    0.12
     harder
    0.12
     higher
    0.11
     Faster
    0.11
    Activations Density 0.101%

    No Known Activations