© 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. 18-LLAMASCOPE-RES-32K
    11. 27021
    Prev
    Next
    INDEX
    Explanations

    phrases where the subject "I" references their experiences or possessions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    osto
    -0.14
    Ùİد
    -0.14
    redients
    -0.14
    ero
    -0.14
    wij
    -0.13
    GY
    -0.13
    \Component
    -0.13
    lady
    -0.13
    ä¸ģ
    -0.13
    ä½ķ
    -0.13
    POSITIVE LOGITS
     situation
    0.17
    lotte
    0.16
    $MESS
    0.16
     situations
    0.15
    /use
    0.15
    .flat
    0.15
     two
    0.15
    å¦Ĥä¸ĭ
    0.15
    gesi
    0.15
    ìľ¨
    0.15
    Activations Density 0.042%

    No Known Activations