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. 130690
    Prev
    Next
    INDEX
    Explanations

    send

    np_max-act · gemini-2.0-flash

    This neuron activates on the imperative verb "put," especially when issuing instructions to put something into a particular state.

    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
    293
    -0.06
    294
    -0.06
     AFP
    -0.06
     visas
    -0.06
    分别
    -0.06
     التي
    -0.06
    ěti
    -0.06
    _exception
    -0.06
    راه
    -0.06
     kanıt
    -0.06
    POSITIVE LOGITS
     Period
    0.07
     Radio
    0.07
    _DEFAULT
    0.06
    =my
    0.06
     khiến
    0.06
     Branch
    0.06
     Bass
    0.06
     Suppose
    0.06
     Auto
    0.06
     My
    0.06
    Activations Density 0.342%

    No Known Activations