Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 81446
    Prev
    Next
    INDEX
    Explanations

    hill

    np_max-act · gemini-2.0-flash

    The neuron flags mentions of elevated terrain—words indicating a location on a hill, bluff, crest, ridge, or similar high ground.

    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
     shar
    -0.07
     witness
    -0.06
    .Param
    -0.06
    (cost
    -0.06
     умови
    -0.06
    webElementProperties
    -0.06
    Setter
    -0.06
    "}),↵
    -0.06
     případech
    -0.06
    validator
    -0.06
    POSITIVE LOGITS
    तर
    0.07
    ursive
    0.07
    差
    0.07
     bekl
    0.06
     sn
    0.06
     Late
    0.06
    ustain
    0.06
    isEqualTo
    0.06
    md
    0.06
    .getExternalStorage
    0.06
    Activations Density 0.015%

    No Known Activations