© 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. 9701
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal and liability issues in business and insurance contexts

    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
    opus
    -0.13
     веÑī
    -0.13
    abor
    -0.13
    оÑĩек
    -0.12
    åŃĶ
    -0.12
    tsx
    -0.12
    æŁĵ
    -0.12
    rych
    -0.12
    BorderStyle
    -0.12
     ÎĶημο
    -0.12
    POSITIVE LOGITS
     insurance
    0.73
     Insurance
    0.70
     insurers
    0.69
     insurer
    0.67
    Insurance
    0.66
    insurance
    0.64
     insured
    0.61
    ä¿ĿéĻ©
    0.53
     Ins
    0.51
     insure
    0.50
    Activations Density 0.355%

    No Known Activations