© 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. 75358
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate comparisons or performance metrics

    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
    oid
    -0.07
     reins
    -0.06
    åħ
    -0.06
    apt
    -0.06
    APT
    -0.06
     present
    -0.06
    antly
    -0.06
     Jehovah
    -0.06
    лаÑĪ
    -0.05
     perch
    -0.05
    POSITIVE LOGITS
     compared
    0.11
     Compare
    0.10
    ï¼Įæ¯Ķ
    0.09
    æ¯Ķ
    0.09
     compares
    0.09
     æ¯Ķ
    0.08
    Compare
    0.08
     compare
    0.08
    compare
    0.08
     Compared
    0.08
    Activations Density 0.044%

    No Known Activations