© 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 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. 59499
    Prev
    Next
    INDEX
    Explanations

    instances of the word "until."

    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
     kit
    -0.06
     once
    -0.06
     during
    -0.05
    uli
    -0.05
     after
    -0.05
     bear
    -0.05
    antar
    -0.05
    aus
    -0.05
    and
    -0.05
     cult
    -0.05
    POSITIVE LOGITS
    Ñħодим
    0.09
    ãĤ´ãĥª
    0.08
     meantime
    0.08
    BUM
    0.08
    Forgery
    0.08
    окÑĢема
    0.08
    GetObject
    0.07
    kün
    0.07
    ÑĩÑĥ
    0.07
    ONTAL
    0.07
    Activations Density 0.004%

    No Known Activations