© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 19710
    Prev
    Next
    INDEX
    Explanations

    terms and phrases indicating actions and states of being related to leadership and social influence

    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
     Firstly
    -0.24
     firstly
    -0.21
    ãģ¾ãģļ
    -0.21
     jedoch
    -0.19
     jednak
    -0.19
     먼ìłĢ
    -0.18
     však
    -0.18
     however
    -0.17
     åħĪ
    -0.17
     first
    -0.17
    POSITIVE LOGITS
     therefore
    0.64
     consequently
    0.60
     accordingly
    0.56
     thereby
    0.51
     thus
    0.47
     hence
    0.42
     Therefore
    0.40
    Therefore
    0.40
     consequ
    0.40
     subsequently
    0.39
    Activations Density 0.489%

    No Known Activations