© 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. 4-RES-SM
    11. 15199
    Prev
    Next
    INDEX
    Explanations

    details about environmental and ecological research involving various species

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    221
    +0.14
    0.8%
    219
    +0.12
    0.7%
    117
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    221
    +0.14
    0.06
    32
    +0.12
    0.06
    50
    +0.12
    0.10
    Negative Logits
     PCI
    -2.08
    PCI
    -1.97
    anyahu
    -1.71
    PLL
    -1.69
    webpack
    -1.68
     Univ
    -1.66
    vue
    -1.65
     Angular
    -1.62
     Chrysler
    -1.62
     Verizon
    -1.62
    POSITIVE LOGITS
     birds
    3.88
     species
    3.63
     bird
    3.61
     insects
    3.58
     Birds
    3.45
     habitat
    3.41
     foraging
    3.38
     Species
    3.37
     larvae
    3.36
     Wildlife
    3.35
    Activations Density 2.950%

    No Known Activations