Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    3. Llama3.1-8B (Base)
    4. Residual Stream
    5. 31-LLAMASCOPE-RES-131K
    6. 43456
    Prev
    Next
    INDEX
    Explanations

    instances of the command or keyword "new" in a programming or document context

    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
     informed
    -0.06
     a
    -0.06
    imore
    -0.05
    uste
    -0.05
    .
    -0.05
    owan
    -0.05
     for
    -0.05
    irsch
    -0.05
     Nolan
    -0.05
     new
    -0.05
    POSITIVE LOGITS
    length
    0.10
    dim
    0.10
    counter
    0.09
    -length
    0.09
    command
    0.09
    save
    0.09
    environment
    0.09
    thought
    0.08
    reff
    0.08
    -counter
    0.08
    Activations Density 0.003%

    No Known Activations