© 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
    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. 54034
    Prev
    Next
    INDEX
    Explanations

    past tense verbs

    np_max-act · gemini-2.0-flash

    The neuron consistently lights up on past-tense or past-participle verbs ending in “-ed.”

    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
    大会
    -0.07
    озможно
    -0.06
     monumental
    -0.06
    丁目
    -0.06
    就在
    -0.06
    最近
    -0.06
    907
    -0.06
    ）の
    -0.06
    луб
    -0.06
    /etc
    -0.06
    POSITIVE LOGITS
     flo
    0.07
     ori
    0.07
     Prediction
    0.07
     Gör
    0.07
     vict
    0.06
    _ct
    0.06
    ateau
    0.06
     β
    0.06
     Macron
    0.06
     mechanism
    0.06
    Activations Density 0.293%

    No Known Activations