© 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. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 26232
    Prev
    Next
    INDEX
    Explanations

    references to parts or segments of larger entities or concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/1-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.1.hook_resid_post
    Hook Layer
    1
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    20
    +0.15
    0.9%
    100
    +0.15
    0.8%
    245
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    20
    +0.15
    0.02
    448
    +0.15
    0.02
    245
    +0.13
    0.01
    Negative Logits
    ĨĴ
    -3.67
    ĸ
    -3.22
    ½
    -3.21
    ↵↵↵
    -3.15
    ↵
    -3.15
       ↵
    -3.15
    -3.15
    ↵↵
    -3.15
    -3.15
    ↵↵
    -3.15
    POSITIVE LOGITS
     thereof
    2.39
    ière
    2.15
     of
    1.94
    isans
    1.88
     sized
    1.85
    enberg
    1.77
    ridge
    1.68
     length
    1.62
    aneous
    1.61
     Of
    1.57
    Activations Density 0.084%

    No Known Activations