© 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. 23812
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to positivity and collaborative efforts

    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
    478
    +0.16
    0.9%
    494
    +0.14
    0.8%
    156
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    494
    +0.16
    0.21
    209
    +0.14
    0.17
    402
    +0.12
    0.12
    Negative Logits
    ģ
    -2.04
    ugu
    -1.91
     respectively
    -1.76
    ungen
    -1.76
    £
    -1.72
    ior
    -1.72
     (,
    -1.70
    ł
    -1.61
    Ģ
    -1.58
    ung
    -1.58
    POSITIVE LOGITS
    /
    1.45
     tighter
    1.43
     peer
    1.40
    quot
    1.34
     mediation
    1.34
    scriptions
    1.31
    rog
    1.30
    rogens
    1.29
     Mo
    1.29
     other
    1.26
    Activations Density 1.226%

    No Known Activations