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. 14-LLAMASCOPE-RES-32K
    6. 10972
    Prev
    Next
    INDEX
    Explanations

    descriptions of visual and emotional complexity in film scenes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.14.hook_resid_post
    Hook Layer
    14
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    64
    -0.14
    WR
    -0.14
    (typeof
    -0.14
    venues
    -0.14
    umin
    -0.14
     Brad
    -0.13
    PI
    -0.13
     æķħ
    -0.13
    396
    -0.13
     Ware
    -0.13
    POSITIVE LOGITS
    USTER
    0.17
    .scalablytyped
    0.16
    ISIBLE
    0.15
    killer
    0.14
    flash
    0.14
    iser
    0.14
     Král
    0.14
    zimmer
    0.14
    integral
    0.14
    ÑĤÑĮÑģÑı
    0.14
    Activations Density 0.086%

    No Known Activations