© 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. 22-LLAMASCOPE-RES-32K
    11. 1658
    Prev
    Next
    INDEX
    Explanations

    terms and concepts related to taxes and financial security

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    adm
    -0.14
    Ñĥмов
    -0.14
    emoc
    -0.13
     MetroFramework
    -0.13
     Pendant
    -0.13
    alis
    -0.13
    γÏĩ
    -0.13
     display
    -0.13
     нак
    -0.13
    FromClass
    -0.13
    POSITIVE LOGITS
     tax
    0.44
     Tax
    0.40
    Tax
    0.35
    tax
    0.35
    _tax
    0.33
    -tax
    0.33
     IRS
    0.33
     TAX
    0.33
    .tax
    0.33
     taxes
    0.30
    Activations Density 0.091%

    No Known Activations