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. 30-LLAMASCOPE-RES-32K
    6. 17824
    Prev
    Next
    INDEX
    Explanations

    essential verbs and conjunctions that contribute to sentence structure

    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
     завиÑģим
    -0.21
     докÑĥм
    -0.16
    psilon
    -0.15
     недел
    -0.15
     ÑĤабли
    -0.14
    amu
    -0.14
    .opend
    -0.13
    pha
    -0.13
    ÙĥÙħ
    -0.13
    ÑĢониÑĩеÑģ
    -0.13
    POSITIVE LOGITS
     поÑıви
    0.21
     оно
    0.20
     бÑĭло
    0.19
     бÑĭла
    0.19
     ÑįÑĤа
    0.19
     бÑĭли
    0.18
     некоÑĤоÑĢÑĭе
    0.18
    appe
    0.17
     поÑıв
    0.17
     οι
    0.16
    Activations Density 0.052%

    No Known Activations