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. 25-LLAMASCOPE-RES-32K
    6. 16450
    Prev
    Next
    INDEX
    Explanations

    themes of emotional struggle and interpersonal relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    pose
    -0.16
    )prepare
    -0.15
    azard
    -0.15
    urrection
    -0.14
     behold
    -0.14
    ague
    -0.14
    apl
    -0.14
     ФедеÑĢалÑĮ
    -0.14
    ignet
    -0.14
    伸
    -0.13
    POSITIVE LOGITS
     internal
    0.23
     compartment
    0.21
     experience
    0.18
    internal
    0.17
     ideal
    0.17
    experience
    0.17
     feel
    0.16
    æĦŁãģĺ
    0.16
     become
    0.15
     gloss
    0.15
    Activations Density 0.793%

    No Known Activations