© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 129793
    Prev
    Next
    INDEX
    Explanations

    Full

    np_max-act · gemini-2.0-flash

    The neuron consistently fires on the token “Full” (and lowercase “full”), as used in headings or labels like “Full Text,” “Full Print Edition,” and similar metadata markers.

    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
     accident
    -0.09
     ai
    -0.08
    ереж
    -0.07
    GameObject
    -0.07
     arbit
    -0.07
     nec
    -0.07
    Where
    -0.07
    azi
    -0.07
    (IT
    -0.07
     achie
    -0.07
    POSITIVE LOGITS
     full
    0.16
     Full
    0.15
     FULL
    0.13
    Full
    0.12
     fully
    0.11
    .Full
    0.11
    FULL
    0.10
    full
    0.10
     Fuller
    0.10
    .full
    0.10
    Activations Density 0.029%

    No Known Activations