© 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. 28-LLAMASCOPE-RES-32K
    11. 23528
    Prev
    Next
    INDEX
    Explanations

    phrases related to dedication and focus on specific activities or groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.28.hook_resid_post
    Hook Layer
    28
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ãĥ¼ãĥķ
    -0.15
    onte
    -0.14
    IDS
    -0.14
     readily
    -0.13
     Practical
    -0.13
    atak
    -0.13
    ãģ«ãģ¨
    -0.13
    ÃĶNG
    -0.13
    ķìĿ¸
    -0.13
    etro
    -0.13
    POSITIVE LOGITS
     entirely
    0.46
     exclusive
    0.44
     exclus
    0.41
     Sole
    0.40
     exclusively
    0.40
    sole
    0.40
     sole
    0.39
     solely
    0.38
     strictly
    0.37
     completely
    0.35
    Activations Density 0.229%

    No Known Activations