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. 86236
    Prev
    Next
    INDEX
    Explanations

    experience

    np_max-act · gemini-2.0-flash

    The neuron never activates on any content—it isn’t detecting any pattern (i.e. it finds nothing).

    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
     передбач
    -0.06
     तक
    -0.06
    .score
    -0.06
     ALS
    -0.06
    igenous
    -0.06
    .window
    -0.06
    	boost
    -0.06
     مرد
    -0.06
     визнач
    -0.06
     submar
    -0.06
    POSITIVE LOGITS
     Kill
    0.07
    _instance
    0.07
    Problem
    0.07
    Fl
    0.07
    author
    0.06
    IGHT
    0.06
    ->{'
    0.06
    توان
    0.06
    bt
    0.06
     const
    0.06
    Activations Density 0.002%

    No Known Activations