© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 13419
    Prev
    Next
    INDEX
    Explanations

    _import

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on the token “_import,” i.e. the “import” part of Python `from __future__ import …` statements.

    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
     productService
    -0.06
    >';↵↵
    -0.06
    Ru
    -0.06
     maneu
    -0.06
    ตรง
    -0.06
     fmt
    -0.06
     radians
    -0.06
    PLETED
    -0.06
     данные
    -0.06
    ляє
    -0.06
    POSITIVE LOGITS
    _import
    0.08
    pai
    0.06
     pinch
    0.06
    ibaba
    0.06
    +l
    0.06
     двор
    0.06
    スコ
    0.06
    _j
    0.06
     Emp
    0.06
    ConfigureAwait
    0.06
    Activations Density 0.000%

    No Known Activations