© 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. 4-LLAMASCOPE-RES-32K
    11. 14891
    Prev
    Next
    INDEX
    Explanations

    themes related to difficulty and simplicity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Rack
    -0.17
    agra
    -0.17
     sink
    -0.15
    Fixed
    -0.14
     Fixed
    -0.14
     rack
    -0.14
     Sir
    -0.13
    ç¬Ķ
    -0.13
    .locals
    -0.13
    441
    -0.13
    POSITIVE LOGITS
    iox
    0.15
    ùy
    0.15
    jsonp
    0.15
     complexity
    0.15
    лÑıн
    0.15
    /exp
    0.15
    ntax
    0.14
    ughs
    0.14
    WithString
    0.14
    dep
    0.14
    Activations Density 0.231%

    No Known Activations