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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 24-LLAMASCOPE-RES-32K
    11. 18180
    Prev
    Next
    INDEX
    Explanations

    contractions and negations in statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Ø®ÛĮ
    -0.16
    361
    -0.15
    fuck
    -0.15
     обÑģÑĤ
    -0.15
     Fucking
    -0.15
    ITTE
    -0.14
    .metro
    -0.14
    TestFixture
    -0.14
    530
    -0.14
     nig
    -0.14
    POSITIVE LOGITS
    .uc
    0.17
     Dil
    0.17
    Å¡t
    0.16
    andas
    0.15
     Pur
    0.15
    رÙĪد
    0.14
     Ultimate
    0.14
     hr
    0.13
    itional
    0.13
    azing
    0.13
    Activations Density 0.001%

    No Known Activations