Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 84516
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act · gemini-2.0-flash

    The neuron is strongly activated by the word “of” (i.e. by prepositional “of” phrases).

    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
     Bạn
    -0.07
     рань
    -0.07
     there
    -0.07
     northwest
    -0.07
    내
    -0.07
     tracker
    -0.07
    you
    -0.07
    -navigation
    -0.06
     southeast
    -0.06
     بخشی
    -0.06
    POSITIVE LOGITS
     of
    0.30
     OF
    0.25
     Of
    0.24
    of
    0.21
    -of
    0.21
    Of
    0.19
    _of
    0.18
    _Of
    0.17
    OF
    0.15
    _OF
    0.15
    Activations Density 1.876%

    No Known Activations