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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 84409
    Prev
    Next
    INDEX
    Explanations

    way

    np_max-act · gemini-2.0-flash

    This neuron mainly fires on the token “to,” i.e. it detects the infinitive marker “to” in the text.

    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
    genic
    -0.07
     gen
    -0.07
    Berry
    -0.07
     arter
    -0.06
     Essential
    -0.06
     Twitter
    -0.06
     particular
    -0.06
    _free
    -0.06
     František
    -0.06
     [/
    -0.06
    POSITIVE LOGITS
     розповід
    0.07
     checker
    0.06
    TabControl
    0.06
    说
    0.06
     Verb
    0.06
    _User
    0.06
    Block
    0.06
    地说
    0.06
    ilogue
    0.06
     σχέ
    0.06
    Activations Density 0.038%

    No Known Activations