© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 130027
    Prev
    Next
    INDEX
    Explanations

    neural networks

    np_max-act · gemini-2.0-flash

    explanations related to neuronal behavior in large language models.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    This neuron activates on code‐related tokens, especially those referring to retrieving or indexing model activations (e.g., "activations", "layer", "layers", "`layer_index`") in a PyTorch context.

    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
     stor
    -0.07
    ENTICATION
    -0.07
     railway
    -0.07
    _IOCTL
    -0.06
    ValuePair
    -0.06
    ulations
    -0.06
     Cumhurbaş
    -0.06
    edores
    -0.06
    Drivers
    -0.06
    partials
    -0.06
    POSITIVE LOGITS
    cname
    0.06
     Using
    0.06
    ٫
    0.06
    di
    0.06
     допомаг
    0.06
     ];
↵
    0.06
    нув
    0.06
    flip
    0.06
    ofi
    0.05
     Progress
    0.05
    Activations Density 0.004%

    No Known Activations