© 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. 125147
    Prev
    Next
    INDEX
    Explanations

    abroad

    np_max-act · gemini-2.0-flash

    The neuron activates on technical/troubleshooting terms—especially around “error,” “information,” and “maintenance”—identifying log-style or system-maintenance language.

    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
    predictions
    -0.07
    LANGUAGE
    -0.07
    Why
    -0.07
     goats
    -0.07
     pans
    -0.07
     showers
    -0.07
     pops
    -0.07
    �
    -0.07
     GOOD
    -0.07
    pra
    -0.07
    POSITIVE LOGITS
    _IDENTIFIER
    0.06
    _PLAYER
    0.06
    ounded
    0.06
    ematic
    0.06
    \Has
    0.06
     захисту
    0.06
     Datensch
    0.05
    	logrus
    0.05
     ÜNİ
    0.05
     ไทย
    0.05
    Activations Density 0.673%

    No Known Activations