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. 85843
    Prev
    Next
    INDEX
    Explanations

    Programming questions and answers

    np_max-act · gemini-2.0-flash

    This neuron lights up on modal and hedging expressions (words like would, could, maybe, possible, will) that signal conditional or potential actions.

    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
     tower
    -0.07
    gems
    -0.06
     migraine
    -0.06
     admits
    -0.06
     Fog
    -0.06
     πρά
    -0.06
    -ie
    -0.06
     ducks
    -0.06
    _ru
    -0.06
    ैस
    -0.06
    POSITIVE LOGITS
    ocking
    0.07
    othermal
    0.07
    release
    0.06
    onitor
    0.06
     mut
    0.06
    CL
    0.06
     recover
    0.06
     Kentucky
    0.06
    τών
    0.06
     anonym
    0.06
    Activations Density 0.043%

    No Known Activations