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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 126761
    Prev
    Next
    INDEX
    Explanations

    repetitive or exaggerated vowel and consonant patterns in words

    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
    ocity
    -0.08
    eyle
    -0.08
    yš
    -0.07
    iego
    -0.07
     eskort
    -0.07
    天åłĤ
    -0.07
    .scalablytyped
    -0.07
    ÑĢÑĥз
    -0.07
     ëª
    -0.07
    rail
    -0.07
    POSITIVE LOGITS
     ur
    0.07
    etheless
    0.06
     just
    0.06
     fed
    0.06
     sp
    0.06
     possibly
    0.06
     yyyy
    0.05
     ST
    0.05
     om
    0.05
    à¥įà¤
    0.05
    Activations Density 0.011%

    No Known Activations