© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 5-LLAMASCOPE-RES-32K
    11. 23744
    Prev
    Next
    INDEX
    Explanations

    sections of code or structured programming comments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.5.hook_resid_post
    Hook Layer
    5
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    polator
    -0.15
    anc
    -0.14
    ÅĻeb
    -0.14
    angan
    -0.13
    за
    -0.13
    ritis
    -0.13
    ordo
    -0.13
     reddit
    -0.13
     doors
    -0.13
    .hd
    -0.13
    POSITIVE LOGITS
    erce
    0.17
    761
    0.16
    ลà¸ĩ
    0.15
    kate
    0.15
    $MESS
    0.15
     Kauf
    0.14
    amak
    0.14
    ãĥŃãĥ¼
    0.14
    adero
    0.14
    ãĥ¼ãĤ¯
    0.14
    Activations Density 0.029%

    No Known Activations