© 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. 43501
    Prev
    Next
    INDEX
    Explanations

    Garbled text

    np_max-act · gemini-2.0-flash

    The neuron flags non-natural‐language tokens such as punctuation‐only items, numeric strings, and uppercase alphanumeric “code” fragments rather than ordinary words.

    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
    ِه
    -0.07
     đảng
    -0.06
     mest
    -0.06
    ày
    -0.06
    -men
    -0.06
    ่าอ
    -0.06
    	Con
    -0.06
     문자
    -0.06
    +s
    -0.06
    Fax
    -0.06
    POSITIVE LOGITS
    /close
    0.07
    生物
    0.07
    prec
    0.06
    (close
    0.06
    рап
    0.06
    oracle
    0.06
    Exact
    0.06
    .sprites
    0.06
    heap
    0.06
    柔
    0.06
    Activations Density 0.032%

    No Known Activations