© 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. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 29866
    Prev
    Next
    INDEX
    Explanations

    terms related to vastness or extent, especially in a geographical or community context

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    362
    +0.17
    1.0%
    156
    +0.13
    0.7%
    194
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.17
    0.03
    362
    +0.13
    0.03
    298
    +0.12
    0.02
    Negative Logits
     altogether
    -1.84
     aloud
    -1.72
    oscopic
    -1.58
    itian
    -1.54
    ilent
    -1.52
    )$$
    -1.52
     equiv
    -1.48
    ococ
    -1.48
    itious
    -1.48
     arbitrary
    -1.48
    POSITIVE LOGITS
    bank
    1.80
    view
    1.71
     magazine
    1.54
     view
    1.51
     goals
    1.50
    à°¿
    1.49
     Savings
    1.48
     magazines
    1.47
    orld
    1.46
    velopment
    1.45
    Activations Density 0.146%

    No Known Activations