© 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. 77382
    Prev
    Next
    INDEX
    Explanations

    negative values or references to losses in contexts such as scores 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
    uble
    -0.07
    ัมà¸ŀ
    -0.07
    ship
    -0.07
    å³
    -0.06
    inks
    -0.06
    anke
    -0.06
    ãĥĩãĤ£ãĤ¢
    -0.06
    á»ĵng
    -0.06
    odash
    -0.06
    lz
    -0.06
    POSITIVE LOGITS
    _MACRO
    0.07
    aci
    0.06
    osa
    0.06
    auen
    0.06
    509
    0.05
    543
    0.05
     Ala
    0.05
    atrix
    0.05
    umm
    0.05
    opsy
    0.05
    Activations Density 0.001%

    No Known Activations