© 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. 30-LLAMASCOPE-RES-32K
    11. 11513
    Prev
    Next
    INDEX
    Explanations

    phrases that emphasize innovation and discovery of new concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.30.hook_resid_post
    Hook Layer
    30
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ätz
    -0.15
     Verfüg
    -0.15
    ubic
    -0.15
    ãĤ¤ãĤº
    -0.14
    Millis
    -0.14
    htable
    -0.14
    quence
    -0.14
    509
    -0.14
    ampus
    -0.14
    ulerAngles
    -0.14
    POSITIVE LOGITS
     hor
    0.36
     front
    0.29
     directions
    0.26
     Hor
    0.26
    hor
    0.24
     ideas
    0.23
     ways
    0.23
    front
    0.23
     uses
    0.23
     Front
    0.22
    Activations Density 0.087%

    No Known Activations