© 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. 125888
    Prev
    Next
    INDEX
    Explanations

    quotation marks and speech indicators in text

    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
    å²
    -0.07
    ะà¹ģ
    -0.07
     Dün
    -0.07
    ooth
    -0.07
    dden
    -0.07
    Ñĸнг
    -0.07
    fon
    -0.07
    orc
    -0.07
    _Zero
    -0.07
    urum
    -0.07
    POSITIVE LOGITS
     because
    0.07
     without
    0.07
     WITHOUT
    0.06
    ets
    0.06
     Because
    0.06
    ffect
    0.06
     whose
    0.06
    <decltype
    0.06
    ecause
    0.06
    obel
    0.06
    Activations Density 0.020%

    No Known Activations