© 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. 16-LLAMASCOPE-RES-32K
    11. 8674
    Prev
    Next
    INDEX
    Explanations

    monetary amounts and details related to fees and admissions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    iro
    -0.16
     corner
    -0.15
    athers
    -0.15
    awe
    -0.14
    budget
    -0.14
     hom
    -0.14
     Budget
    -0.14
     Expansion
    -0.13
     Term
    -0.13
    pets
    -0.13
    POSITIVE LOGITS
    enco
    0.16
    aise
    0.16
    CoreApplication
    0.15
    DMI
    0.15
    CRET
    0.15
    rena
    0.15
    -entry
    0.15
     redistributed
    0.15
     admission
    0.15
     Admission
    0.15
    Activations Density 0.041%

    No Known Activations