© 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. 30099
    Prev
    Next
    INDEX
    Explanations

    iod

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of “iod” or “ization,” i.e. it detects mentions of iodine/iodization.

    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
     projection
    -0.07
    _LOAD
    -0.07
    .AUTH
    -0.07
    _prov
    -0.07
    _paid
    -0.07
    バイ
    -0.06
     randomNumber
    -0.06
     ATH
    -0.06
    ernaut
    -0.06
     faculties
    -0.06
    POSITIVE LOGITS
     şi
    0.06
    ›
    0.06
     lucrative
    0.06
     iod
    0.06
     etiquette
    0.06
     biçim
    0.06
     lic
    0.06
     그가
    0.06
     Intelligence
    0.05
    ินด
    0.05
    Activations Density 0.001%

    No Known Activations