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. 81402
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act · gemini-2.0-flash

    This neuron specializes in spotting the names of chemical process steps—gerund or nominalized verbs like “Preparation,” “Chlorination,” “Reduction,” “Dehydration,” and “Purification.”

    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
     spinning
    -0.07
    应该
    -0.07
     проти
    -0.06
    Disposed
    -0.06
    _)
    -0.06
    alaria
    -0.06
    WHAT
    -0.06
    这样
    -0.06
     telah
    -0.06
     whispered
    -0.06
    POSITIVE LOGITS
    aleza
    0.07
     Good
    0.06
     %=
    0.06
    _property
    0.06
     PTR
    0.06
     licz
    0.06
     roc
    0.06
     دقی
    0.06
    :'',↵
    0.06
    _ptrs
    0.06
    Activations Density 0.019%

    No Known Activations