© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 110358
    Prev
    Next
    INDEX
    Explanations

    Number sequences

    np_max-act · gemini-2.0-flash

    This neuron detects numeric tokens in “What comes next” sequence puzzles, especially the numbers at the ends of comma-separated lists.

    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
     urlparse
    -0.07
    \V
    -0.06
    Purpose
    -0.06
     Shaun
    -0.06
     Main
    -0.06
    weeney
    -0.06
    Te
    -0.06
     curry
    -0.06
    SerializeField
    -0.06
    _Bool
    -0.06
    POSITIVE LOGITS
     ماسه
    0.07
    GUILayout
    0.07
     کنند
    0.06
    ีค
    0.06
    aş
    0.06
    =batch
    0.06
     hamm
    0.06
    丽
    0.06
     tracing
    0.06
    _Back
    0.06
    Activations Density 0.053%

    No Known Activations