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. 44145
    Prev
    Next
    INDEX
    Explanations

    phrases indicating communication or exchange of ideas

    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
    zin
    -0.07
    instein
    -0.07
    terior
    -0.07
     permanently
    -0.07
    ensch
    -0.07
    permanent
    -0.07
    ään
    -0.07
    evin
    -0.07
    éĩĩ
    -0.06
    alli
    -0.06
    POSITIVE LOGITS
     abundance
    0.08
     presently
    0.08
     discourse
    0.07
     Sir
    0.06
     upon
    0.06
     desired
    0.06
     cry
    0.06
     bespoke
    0.06
    meth
    0.06
     whispered
    0.06
    Activations Density 0.004%

    No Known Activations