© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 26-LLAMASCOPE-RES-32K
    11. 15080
    Prev
    Next
    INDEX
    Explanations

    technical terms and concepts related to audio and data streams in programming and algorithms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.26.hook_resid_post
    Hook Layer
    26
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     incoming
    -0.16
     Incoming
    -0.16
     imported
    -0.15
     inward
    -0.15
    incoming
    -0.15
    ifstream
    -0.15
    Incoming
    -0.15
    -in
    -0.15
     Ellison
    -0.14
    stdin
    -0.14
    POSITIVE LOGITS
     output
    1.13
    output
    1.01
     Output
    1.00
    Output
    0.94
     outputs
    0.91
    _output
    0.88
    .output
    0.88
    -output
    0.88
     OUTPUT
    0.88
    è¾ĵåĩº
    0.84
    Activations Density 0.135%

    No Known Activations