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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 79988
    Prev
    Next
    INDEX
    Explanations

    phrases related to leaving or remaining elements in a context

    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
    hek
    -0.07
    sled
    -0.07
    ouser
    -0.07
    é¡į
    -0.07
    icate
    -0.07
    ibur
    -0.06
    ostel
    -0.06
    ikal
    -0.06
    row
    -0.06
    iven
    -0.06
    POSITIVE LOGITS
     remaining
    0.14
    åī©
    0.12
    remaining
    0.12
     Remaining
    0.12
    Remaining
    0.11
    çķĻ
    0.09
     remained
    0.09
     only
    0.09
     residual
    0.08
    æ®ĭ
    0.08
    Activations Density 0.032%

    No Known Activations