© 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. 41455
    Prev
    Next
    INDEX
    Explanations

    programming code

    np_max-act · gemini-2.0-flash

    The neuron fires on action‐oriented instruction words—especially forms of “execute/run” and related setup verbs (e.g. executar, instalar, criar)—i.e. tokens that signal the user to perform a command.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    instructions related to executing Jupyter notebooks in specific Python environments.

    oai_token-act-pair · gpt-4o-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
     agenda
    -0.07
    temperature
    -0.07
    ゲーム
    -0.07
    sian
    -0.06
    leveland
    -0.06
    -community
    -0.06
    -training
    -0.06
    arius
    -0.06
    Perhaps
    -0.06
     interactions
    -0.06
    POSITIVE LOGITS
     mohla
    0.07
     SHALL
    0.07
     Puppet
    0.07
     був
    0.07
     Shall
    0.07
    failed
    0.06
     denně
    0.06
     색
    0.06
    sav
    0.06
    ='".
    0.06
    Activations Density 0.017%

    No Known Activations