© 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. 84596
    Prev
    Next
    INDEX
    Explanations

    contractions of the word "not."

    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
    hen
    -0.07
    /or
    -0.07
     no
    -0.07
     olmayan
    -0.07
     not
    -0.07
    -↵↵
    -0.06
    holding
    -0.06
    -↵
    -0.06
    uit
    -0.06
    hop
    -0.06
    POSITIVE LOGITS
     necessarily
    0.13
    amp
    0.10
    quot
    0.10
     quite
    0.09
    ori
    0.09
    ches
    0.09
    theless
    0.09
    ecessarily
    0.09
    oriously
    0.08
    ÌĨ
    0.08
    Activations Density 0.135%

    No Known Activations