© 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. 54086
    Prev
    Next
    INDEX
    Explanations

    commission

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of “commission” (and its derivatives) when referring to official bodies, roles, or acts (e.g. Commission, commissioner, commissioning).

    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
     Hat
    -0.07
     Tat
    -0.07
    2
    -0.07
    attle
    -0.07
    ut
    -0.07
     Kata
    -0.07
     knew
    -0.06
     Ut
    -0.06
    Var
    -0.06
     Tut
    -0.06
    POSITIVE LOGITS
     Commission
    0.15
     commission
    0.11
    Commission
    0.10
     commissions
    0.09
    JC
    0.09
     Cameron
    0.08
     Thompson
    0.08
    CBD
    0.08
    psc
    0.08
     CJ
    0.08
    Activations Density 0.008%

    No Known Activations