© 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. 9-LLAMASCOPE-RES-32K
    11. 26872
    Prev
    Next
    INDEX
    Explanations

    the word "are" and its conjugations in various forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    (s
    -0.15
    coli
    -0.14
     instead
    -0.14
    213
    -0.14
     rank
    -0.13
    oup
    -0.13
    ante
    -0.13
     among
    -0.13
    ätz
    -0.13
    chooser
    -0.13
    POSITIVE LOGITS
    icer
    0.15
    BSITE
    0.15
     COPYING
    0.15
    ãĤ¤ãĥ¤
    0.14
    ubern
    0.14
    isser
    0.14
    akedirs
    0.14
    assen
    0.14
    à¥ģत
    0.14
     дÑĥ
    0.13
    Activations Density 0.064%

    No Known Activations