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. 43864
    Prev
    Next
    INDEX
    Explanations

    phrases related to introductory content or resources

    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
     unsubscribe
    -0.06
    /***/
    -0.06
    íĴ
    -0.06
    opal
    -0.06
     intervene
    -0.06
    687
    -0.06
    _INTR
    -0.06
     resend
    -0.05
     Reported
    -0.05
    esteem
    -0.05
    POSITIVE LOGITS
    ductory
    0.16
    ary
    0.15
     into
    0.15
     to
    0.12
    duct
    0.11
     Into
    0.11
    al
    0.10
    into
    0.10
    _into
    0.10
    /welcome
    0.10
    Activations Density 0.014%

    No Known Activations