Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    3. Llama3.1-8B (Base)
    4. Residual Stream
    5. 21-LLAMASCOPE-RES-32K
    6. 9946
    Prev
    Next
    INDEX
    Explanations

    words and variations related to "cube" and its derivatives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    pst
    -0.16
    .mybatisplus
    -0.16
    ighthouse
    -0.16
    ividual
    -0.16
    ional
    -0.15
    kok
    -0.15
    _nsec
    -0.15
    amage
    -0.14
    å¹ķ
    -0.14
    jte
    -0.14
    POSITIVE LOGITS
     Cub
    0.24
     cub
    0.24
    icle
    0.22
    bing
    0.22
    bage
    0.21
    bies
    0.19
    oulder
    0.18
    oidal
    0.18
    ical
    0.18
    rub
    0.18
    Activations Density 0.006%

    No Known Activations