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

    Neuronpedia

    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. 40024
    Prev
    Next
    INDEX
    Explanations

    specific technical elements and jargon related to algorithms and game mechanics

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

    Two-letter combinations where the second letter is often "T", representing either user identifiers, commentators, or technical abbreviations appearing in various contexts including forum posts, code, and articles.

    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
    hide
    -0.11
    hart
    -0.10
    hand
    -0.10
    hit
    -0.10
    hol
    -0.10
    er
    -0.09
    ho
    -0.09
    halt
    -0.09
    hot
    -0.09
    erb
    -0.09
    POSITIVE LOGITS
    ting
    0.19
    ted
    0.15
    ters
    0.12
    ãģĬãĤĬ
    0.12
    aylor
    0.11
    tings
    0.11
    ropolis
    0.11
    unnel
    0.11
    imestep
    0.11
    iffany
    0.10
    Activations Density 1.115%

    No Known Activations