© 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. 7424
    Prev
    Next
    INDEX
    Explanations

    Time/past events

    np_max-act · gemini-2.0-flash

    The neuron lights up on words and phrases that refer to durations or spans of time (e.g. “years,” “decades,” auxiliaries like “had” in temporal contexts).

    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
     caracter
    -0.06
    Quotes
    -0.06
    039
    -0.06
    anmar
    -0.06
    _off
    -0.06
    _nm
    -0.06
    Template
    -0.06
    _XML
    -0.06
     Dew
    -0.06
     Other
    -0.06
    POSITIVE LOGITS
     топ
    0.07
    ResourceId
    0.07
     dục
    0.06
                                    
    0.06
    ULATION
    0.06
     дав
    0.06
     эф
    0.06
    urdu
    0.06
    ┴
    0.06
    ellation
    0.06
    Activations Density 0.144%

    No Known Activations