© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 27745
    Prev
    Next
    INDEX
    Explanations

    phrases related to software licensing and terms

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    320
    +0.14
    0.8%
    450
    +0.13
    0.7%
    230
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    222
    +0.14
    0.02
    103
    +0.13
    0.02
    230
    +0.10
    0.01
    Negative Logits
    ®
    -1.83
    »¿
    -1.83
    į
    -1.74
    ¢
    -1.74
    ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ
    -1.71
    ¿½
    -1.70
    ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ
    -1.69
      ↵
    -1.62
    -1.62
    ↵↵
    -1.62
    POSITIVE LOGITS
     license
    1.77
     licenses
    1.75
     redist
    1.55
     integer
    1.55
     GPL
    1.51
     referee
    1.46
     freely
    1.44
     lic
    1.42
     Void
    1.42
    icense
    1.40
    Activations Density 0.074%

    No Known Activations