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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 30-LLAMASCOPE-RES-32K
    6. 22426
    Prev
    Next
    INDEX
    Explanations

    descriptions of architectural features and structural details

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.30.hook_resid_post
    Hook Layer
    30
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ieri
    -0.16
     поба
    -0.15
    /github
    -0.14
    MouseEvent
    -0.14
    emean
    -0.14
    yne
    -0.14
    çł²
    -0.13
    agner
    -0.13
    icum
    -0.13
     yacc
    -0.13
    POSITIVE LOGITS
     backlash
    0.19
    lashes
    0.16
     fixation
    0.15
     schemas
    0.15
     mutually
    0.15
     Markus
    0.15
     otp
    0.15
    fix
    0.15
     sighting
    0.15
    lara
    0.15
    Activations Density 0.027%

    No Known Activations