© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 17-LLAMASCOPE-RES-131K
    11. 127818
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate starting or building something from the ground up

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Insecta
    -0.07
    tae
    -0.07
    зм
    -0.07
    ROME
    -0.07
    .scalablytyped
    -0.07
    esModule
    -0.07
     hrom
    -0.07
    OffsetTable
    -0.07
    emailer
    -0.07
     tuz
    -0.07
    POSITIVE LOGITS
     scratch
    0.13
    scratch
    0.09
     Scratch
    0.08
     raw
    0.07
     existing
    0.07
     scratched
    0.06
     information
    0.06
     collected
    0.06
     available
    0.06
    itti
    0.06
    Activations Density 0.046%

    No Known Activations