© 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. 31-LLAMASCOPE-RES-32K
    11. 10425
    Prev
    Next
    INDEX
    Explanations

    attributes and qualities related to experiences and evaluations of various subjects or concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     the
    -0.20
    ëŀij
    -0.16
    -addon
    -0.14
    /from
    -0.14
    -Identifier
    -0.14
     its
    -0.14
    thed
    -0.14
    é§ħå¾ĴæŃ©
    -0.14
    	the
    -0.13
    -msg
    -0.13
    POSITIVE LOGITS
    ,
    0.46
     but
    0.45
     and
    0.44
     yet
    0.42
    -but
    0.41
    -looking
    0.39
     albeit
    0.35
    -y
    0.33
    but
    0.31
    yet
    0.30
    Activations Density 0.724%

    No Known Activations