© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 15-LLAMASCOPE-RES-131K
    11. 86641
    Prev
    Next
    INDEX
    Explanations

    parts of speech related to directives and instructions

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

    Conjunction tokens that link clauses in instructional or informational text, typically introducing purpose, consequence, or continuation after a request or statement.

    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
     PureComponent
    -0.07
     vent
    -0.07
    ablish
    -0.07
    hausen
    -0.07
    ofil
    -0.06
    icom
    -0.06
    alm
    -0.06
    ForKey
    -0.06
    nez
    -0.06
    FAQ
    -0.06
    POSITIVE LOGITS
    ziej
    0.07
    adera
    0.06
     Processing
    0.06
    amage
    0.06
     Nack
    0.06
    ARI
    0.06
    ogenerated
    0.06
    ewood
    0.06
     Fet
    0.06
     Giang
    0.06
    Activations Density 0.043%

    No Known Activations