© 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. 95122
    Prev
    Next
    INDEX
    Explanations

    phrases related to challenges or difficulties faced in various contexts

    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
    rien
    -0.08
    è¢
    -0.08
    arget
    -0.08
    æ§ĺ
    -0.07
    uhl
    -0.07
    /ion
    -0.07
    CCR
    -0.07
    spender
    -0.07
    oret
    -0.07
    inati
    -0.07
    POSITIVE LOGITS
     even
    0.12
    even
    0.10
     almost
    0.10
     sogar
    0.09
    almost
    0.08
     даже
    0.08
     EVEN
    0.08
     actually
    0.08
     Even
    0.07
     actual
    0.07
    Activations Density 0.052%

    No Known Activations