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. 31826
    Prev
    Next
    INDEX
    Explanations

    elements indicating fictional content and disclaimers about the accuracy of information

    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
    alles
    -0.15
    fty
    -0.15
    occo
    -0.15
    utsch
    -0.14
    rary
    -0.14
    ennie
    -0.14
     van
    -0.13
    ateurs
    -0.13
    dl
    -0.13
    akes
    -0.13
    POSITIVE LOGITS
    undle
    0.14
    /gtest
    0.14
    _iff
    0.14
    ublic
    0.13
    ãĥ³ãĥĪ
    0.13
    à¥ĩà¤Ĺ
    0.13
     _$
    0.13
     NONINFRINGEMENT
    0.13
    ãĥ³ãĥģ
    0.13
    izar
    0.12
    Activations Density 0.020%

    No Known Activations