© 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. 126690
    Prev
    Next
    INDEX
    Explanations

    negative aspects

    np_max-act · gemini-2.0-flash

    The neuron activates on language pointing out defects, disadvantages, or shortcomings—i.e. negative evaluations of prior work.

    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
     itm
    -0.08
     ache
    -0.07
    lection
    -0.07
    _Login
    -0.07
     flower
    -0.06
     Hồ
    -0.06
    .login
    -0.06
     Winning
    -0.06
     methods
    -0.06
    δώ
    -0.06
    POSITIVE LOGITS
     хими
    0.07
    pyx
    0.07
     intermedi
    0.06
     OutlineInputBorder
    0.06
    ignore
    0.06
    582
    0.06
    glyphicon
    0.06
     FirebaseDatabase
    0.06
     IService
    0.06
     $#
    0.06
    Activations Density 0.026%

    No Known Activations