© 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. 125269
    Prev
    Next
    INDEX
    Explanations

    JSON code

    np_max-act · gemini-2.0-flash

    JSON structures and their properties.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    This neuron identifies JSON object property names (i.e. the keys in JSON-like dotted or bracketed notation).

    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
    _ph
    -0.06
     біл
    -0.06
    (buf
    -0.06
    	clock
    -0.06
    由于
    -0.06
     rage
    -0.06
    μφωνα
    -0.06
     бал
    -0.06
    ，但
    -0.06
    -values
    -0.06
    POSITIVE LOGITS
     Disposable
    0.06
     landlords
    0.06
     indicated
    0.06
    <i
    0.06
    .reactivex
    0.06
     efter
    0.06
     mainBundle
    0.06
    .creator
    0.06
    bine
    0.06
    .NET
    0.06
    Activations Density 0.043%

    No Known Activations