© 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. 16215
    Prev
    Next
    INDEX
    Explanations

    phrases that denote "best" or "top" rankings of various subjects

    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
     наб
    -0.08
    imore
    -0.07
     Ars
    -0.07
    ÑĢава
    -0.07
    oldt
    -0.07
    onda
    -0.06
    rowable
    -0.06
    ãĤį
    -0.06
    rede
    -0.06
    imedia
    -0.06
    POSITIVE LOGITS
     best
    0.07
    ocr
    0.07
    best
    0.07
    aland
    0.06
     cr
    0.06
    dap
    0.06
    -best
    0.06
     Best
    0.06
    atively
    0.06
     recess
    0.06
    Activations Density 0.005%

    No Known Activations