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. 8357
    Prev
    Next
    INDEX
    Explanations

    occurrences of specific words that contain "ugg" or are related to 'guggenheim'

    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
    bjerg
    -0.17
    Ø¢ÙĨ
    -0.14
    acco
    -0.14
    illis
    -0.14
    annels
    -0.14
    avir
    -0.14
    輪
    -0.14
    æŃ
    -0.14
    _heads
    -0.14
    cams
    -0.13
    POSITIVE LOGITS
    ogui
    0.17
    Normals
    0.15
     Reflex
    0.14
     chaud
    0.14
    anteed
    0.14
     disposition
    0.14
    eid
    0.14
    phant
    0.13
    izza
    0.13
    íĴį
    0.13
    Activations Density 0.060%

    No Known Activations