© 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. 68757
    Prev
    Next
    INDEX
    Explanations

    coefficients and powers

    np_max-act · gemini-2.0-flash

    The neuron responds to numerical constants (digits and numbers) within mathematical expressions.

    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
     deepen
    -0.06
    Nama
    -0.06
    Google
    -0.06
    RICS
    -0.06
     друг
    -0.06
    �
    -0.06
    filtr
    -0.06
     abych
    -0.05
     некотор
    -0.05
     Nexus
    -0.05
    POSITIVE LOGITS
     asserting
    0.07
     thinking
    0.07
     Upon
    0.07
     lawful
    0.07
     justified
    0.07
    ल
    0.07
     assignment
    0.07
    -base
    0.07
     conquest
    0.06
    ["_
    0.06
    Activations Density 0.012%

    No Known Activations