Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 125673
    Prev
    Next
    INDEX
    Explanations

    Romance novels

    np_max-act · gemini-2.0-flash

    The neuron detects first-person Russian pronouns and verb forms (e.g. “Я,” “буду,” “меня”) indicating the speaker’s perspective.

    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
     consequently
    -0.07
    .Root
    -0.06
    陷
    -0.06
    mann
    -0.06
    atır
    -0.06
    description
    -0.06
     이러한
    -0.06
     supra
    -0.06
    Parameter
    -0.06
    )'],↵
    -0.06
    POSITIVE LOGITS
     ifs
    0.07
    056
    0.07
    中文
    0.07
     methodologies
    0.07
     ll
    0.06
    -dem
    0.06
     κο
    0.06
     odpově
    0.06
     Vocabulary
    0.06
     Stations
    0.06
    Activations Density 0.136%

    No Known Activations