© 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. 128625
    Prev
    Next
    INDEX
    Explanations

    financial deficits

    np_max-act · gemini-2.0-flash

    references to financial constraints and budget-related discussions.

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

    This neuron activates on terms related to budgetary figures and financial adjustments, such as deficits, cuts, millage rates, and shortfalls.

    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
     ant
    -0.06
    .MixedReality
    -0.06
    	curr
    -0.06
     retaining
    -0.06
    kiye
    -0.06
     scanners
    -0.06
    iye
    -0.06
     vệ
    -0.06
    _saida
    -0.06
     하지만
    -0.06
    POSITIVE LOGITS
    融
    0.07
     l
    0.06
    �
    0.06
    relevant
    0.06
    _email
    0.06
    006
    0.06
    IDDLE
    0.06
    _Des
    0.06
     ordained
    0.06
    146
    0.06
    Activations Density 0.016%

    No Known Activations