© 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. 119464
    Prev
    Next
    INDEX
    Explanations

    sentences with a high frequency of commas

    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
    nds
    -0.06
    elyn
    -0.06
    jis
    -0.06
    jab
    -0.06
    dit
    -0.06
     mÃł
    -0.06
    gger
    -0.06
    anel
    -0.06
    orem
    -0.06
    ÅĦ
    -0.06
    POSITIVE LOGITS
    echa
    0.08
     ‘
    0.07
     Dr
    0.06
    otec
    0.06
     âľĶ
    0.06
    ãĢģ“
    0.06
     Mr
    0.06
    oling
    0.06
    athe
    0.06
     deb
    0.06
    Activations Density 0.025%

    No Known Activations