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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 1229
    Prev
    Next
    INDEX
    Explanations

    R

    np_max-act · gemini-2.0-flash

    The neuron fires on numeric measurements—especially decimal‐formatted numbers—such as crystal lattice parameters and volumes.

    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
     Wolves
    -0.07
    ्थन
    -0.07
     войны
    -0.06
     trails
    -0.06
    _clicked
    -0.06
    _any
    -0.06
     victims
    -0.06
     erfahren
    -0.06
    ством
    -0.06
     Worship
    -0.06
    POSITIVE LOGITS
    	struct
    0.06
    обрет
    0.06
     adjunct
    0.06
     projected
    0.06
    -motion
    0.06
     adopted
    0.06
    stell
    0.06
     oxidative
    0.06
    (created
    0.06
    Anti
    0.06
    Activations Density 0.000%

    No Known Activations