© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 56121
    Prev
    Next
    INDEX
    Explanations

    specific tokens, possibly indicative of coding or formatting syntax

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Parts of proper names or technical identifiers, often appearing in formal contexts like academic citations, technical documentation, or news reporting.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    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
    elman
    -0.07
    icut
    -0.07
    ulative
    -0.07
    ÏįÏĦε
    -0.07
    Prince
    -0.06
    -Agent
    -0.06
    >NN
    -0.06
    CUS
    -0.06
    735
    -0.06
    ijkstra
    -0.06
    POSITIVE LOGITS
    quo
    0.08
    (Int
    0.07
     Zucker
    0.07
    spell
    0.06
    aret
    0.06
    ronic
    0.06
     Murdoch
    0.06
     Natal
    0.06
    eÅŁit
    0.06
    oric
    0.06
    Activations Density 0.098%

    No Known Activations