© 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. 126079
    Prev
    Next
    INDEX
    Explanations

    originally

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens representing floating‐point decimals (fractional numbers).

    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
    Filters
    -0.06
    _VIS
    -0.06
     aliqua
    -0.06
     Patient
    -0.06
    医疗
    -0.06
    EndInit
    -0.06
     sails
    -0.06
     superClass
    -0.06
    -type
    -0.06
     перем
    -0.06
    POSITIVE LOGITS
     originally
    0.12
     Originally
    0.10
     initially
    0.09
    Originally
    0.07
    indy
    0.07
    lında
    0.07
     &&
    0.07
     Saddam
    0.07
     Анд
    0.07
     initial
    0.07
    Activations Density 0.008%

    No Known Activations