© 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. 21-LLAMASCOPE-RES-32K
    11. 7769
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate conditions or criteria related to an agreement or contract

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     regards
    -0.15
    external
    -0.15
    933
    -0.15
    rof
    -0.15
    097
    -0.15
    idis
    -0.14
    ikal
    -0.13
     Ness
    -0.13
     Verfügung
    -0.13
    976
    -0.13
    POSITIVE LOGITS
     behalf
    0.28
     basis
    0.25
    basis
    0.24
     grounds
    0.22
     condition
    0.22
    beh
    0.20
     Basis
    0.19
    condition
    0.17
    _basis
    0.16
    asis
    0.16
    Activations Density 0.087%

    No Known Activations