© 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. 2-RES-SM
    11. 14153
    Prev
    Next
    INDEX
    Explanations

    references to testing and its related activities

    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/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
    Neuron Alignment
    Index
    Value
    % of L₁
    281
    +0.11
    0.6%
    339
    +0.11
    0.6%
    484
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    124
    +0.11
    0.03
    189
    +0.11
    0.02
    465
    +0.11
    0.03
    Negative Logits
    Ļª
    -3.76
    Ĵ
    -3.29
    ¯
    -3.28
    §
    -3.20
    IJ
    -3.10
    ĸ
    -3.06
    ĨĴ
    -2.86
    ij
    -2.84
    Ń
    -2.84
    ©
    -2.80
    POSITIVE LOGITS
    ivities
    1.86
    ifies
    1.83
     accuracy
    1.72
    osterone
    1.70
     kits
    1.62
     automation
    1.61
     amp
    1.61
     accur
    1.60
     testing
    1.58
    icles
    1.57
    Activations Density 0.107%

    No Known Activations