© 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. 41411
    Prev
    Next
    INDEX
    Explanations

    complete sentences

    np_max-act · gemini-2.0-flash

    This neuron detects caveats and hedging language—tokens like "does," "not," "have," "involve," "enough," and similar words used in disclaimers or qualifications.

    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
    鸡
    -0.07
    localhost
    -0.06
    itest
    -0.06
    organizations
    -0.06
     kullanılır
    -0.06
     니
    -0.06
     Nik
    -0.06
    杜
    -0.06
    ():↵
    -0.06
     mesma
    -0.06
    POSITIVE LOGITS
    ाइन
    0.07
    +',
    0.07
    acích
    0.07
    ANC
    0.07
     exponentially
    0.06
     대표
    0.06
    \Order
    0.06
    (Binary
    0.06
    _Mouse
    0.06
     semaine
    0.06
    Activations Density 0.160%

    No Known Activations