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. 26610
    Prev
    Next
    INDEX
    Explanations

    7

    np_max-act · gemini-2.0-flash

    The neuron is looking for floating‐point numeric constants (decimal numbers) in math formulas.

    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
    ddie
    -0.06
    .Car
    -0.06
    �
    -0.06
    .twimg
    -0.06
     cade
    -0.06
     four
    -0.06
    ştır
    -0.06
    _SETTING
    -0.06
     metic
    -0.06
     tslib
    -0.06
    POSITIVE LOGITS
    DATABASE
    0.07
     Mt
    0.07
     صنع
    0.07
    elist
    0.07
     Совет
    0.07
    LBL
    0.06
    َن
    0.06
     Obt
    0.06
    Mt
    0.06
    нт
    0.06
    Activations Density 0.001%

    No Known Activations