© 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. 85716
    Prev
    Next
    INDEX
    Explanations

    numbers

    np_max-act · gemini-2.0-flash

    The neuron detects chemical oxidation‐state notations, i.e. superscripted numbers (often with +/–) indicating an element’s valence.

    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
     yeter
    -0.07
    LANG
    -0.06
    .baidu
    -0.06
     chẳng
    -0.06
    เช
    -0.06
    _backup
    -0.06
     indiscrim
    -0.06
    _DST
    -0.06
     глиб
    -0.06
    unread
    -0.06
    POSITIVE LOGITS
    '],↵
    0.07
     Rooney
    0.07
    strtolower
    0.07
     plank
    0.06
     силы
    0.06
    	global
    0.06
     presenting
    0.06
    _linear
    0.06
     erect
    0.06
     Relevant
    0.06
    Activations Density 0.006%

    No Known Activations