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. 25-LLAMASCOPE-RES-32K
    6. 22587
    Prev
    Next
    INDEX
    Explanations

    statistical references and analyses related to economic indicators and household data

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ROC
    -0.15
     predicted
    -0.14
    Predict
    -0.14
     Et
    -0.14
     bufio
    -0.14
     predicts
    -0.14
     sess
    -0.13
     predicting
    -0.13
     Predict
    -0.13
    graded
    -0.13
    POSITIVE LOGITS
     Statistics
    0.23
    .Stat
    0.22
     statist
    0.22
     Coverage
    0.21
     Harmon
    0.21
    Stat
    0.21
     Stats
    0.21
    coverage
    0.20
    STAT
    0.20
    Coverage
    0.20
    Activations Density 0.052%

    No Known Activations