© 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. 850
    Prev
    Next
    INDEX
    Explanations

    becoming obsolete

    np_max-act · gemini-2.0-flash

    The neuron activates on words and word parts referring to obsolescence—that is, terms indicating something is outdated or no longer in use.

    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
    client
    -0.07
    Root
    -0.06
     MetroFramework
    -0.06
    プレ
    -0.06
    Krist
    -0.06
    Ubergraph
    -0.06
    _ATTRIBUTES
    -0.06
     soud
    -0.06
     ESP
    -0.06
     остав
    -0.06
    POSITIVE LOGITS
     obsolete
    0.09
     outdated
    0.08
    soles
    0.07
    ???
    0.07
    modity
    0.06
     This
    0.06
     dropdown
    0.06
    thood
    0.06
    bral
    0.06
     menggunakan
    0.06
    Activations Density 0.012%

    No Known Activations