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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 84405
    Prev
    Next
    INDEX
    Explanations

    software changes/lists

    np_max-act · gemini-2.0-flash

    This neuron primarily detects floating‐point or decimal numeric tokens (numbers with fractional parts).

    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
     NSLog
    -0.07
     handset
    -0.07
     attributable
    -0.07
     �
    -0.07
     rawData
    -0.07
    mind
    -0.06
    ">'+↵
    -0.06
    ソン
    -0.06
     �
    -0.06
    arehouse
    -0.06
    POSITIVE LOGITS
    velocity
    0.07
    ็นท
    0.07
    ınızda
    0.06
    	ast
    0.06
    Outlined
    0.06
     avid
    0.06
     electr
    0.06
    oub
    0.06
     hydrogen
    0.06
     housing
    0.06
    Activations Density 0.262%

    No Known Activations