Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    3. Llama3.1-8B (Base)
    4. Residual Stream
    5. 11-LLAMASCOPE-RES-32K
    6. 964
    Prev
    Next
    INDEX
    Explanations

    occurrences of the definite article "the" and other specific contextual clues within technical or procedural content

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.11.hook_resid_post
    Hook Layer
    11
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    edom
    -0.17
    bish
    -0.15
    ushman
    -0.15
    orning
    -0.14
    ofile
    -0.14
    adle
    -0.14
    inh
    -0.14
    ematic
    -0.14
    ulace
    -0.13
    ymb
    -0.13
    POSITIVE LOGITS
    LookAndFeel
    0.16
    inded
    0.16
    hai
    0.14
    λÏĮγ
    0.14
    acular
    0.14
    vre
    0.14
    olar
    0.14
     Weed
    0.14
    tha
    0.13
    ovy
    0.13
    Activations Density 0.009%

    No Known Activations