© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 126404
    Prev
    Next
    INDEX
    Explanations

    prepositions of location

    np_max-act · gemini-2.0-flash

    narratives about personal transformations and adaptations in challenging situations.

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

    The neuron detects ominous atmospheric action words that convey something enveloping or looming (e.g. “descended,” “swallowed up,” “settled over,” “loomed”).

    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
     Studios
    -0.07
    仓
    -0.07
     Nacional
    -0.07
     تف
    -0.06
    rů
    -0.06
     Dest
    -0.06
    ově
    -0.06
    manuel
    -0.06
    _alt
    -0.06
    Tutorial
    -0.06
    POSITIVE LOGITS
    (super
    0.07
     خود
    0.07
    طب
    0.06
    abilece
    0.06
     trùng
    0.06
    (lon
    0.06
     konusu
    0.06
     здійснення
    0.06
     خانواده
    0.06
     TRACK
    0.06
    Activations Density 0.041%

    No Known Activations