© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 125107
    Prev
    Next
    INDEX
    Explanations

    Meditation

    np_max-act · gemini-2.0-flash

    This neuron activates on transliterated Sanskrit terms (often with macrons), i.e. words like “dhyāna” and “bhāvanā.”

    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
    .rank
    -0.06
     mpz
    -0.06
     dumped
    -0.06
    .FindGameObjectWithTag
    -0.06
     ApplicationRecord
    -0.06
    Reduc
    -0.06
    線
    -0.06
    /kg
    -0.06
    -th
    -0.06
    غيل
    -0.06
    POSITIVE LOGITS
     воспал
    0.07
     farming
    0.07
     INVALID
    0.07
     endeavors
    0.07
     bisexual
    0.06
    ές
    0.06
    !
    0.06
    innerText
    0.06
     textStatus
    0.06
    ็บไซต
    0.06
    Activations Density 0.011%

    No Known Activations