© 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. 4-RES-SM
    11. 13721
    Prev
    Next
    INDEX
    Explanations

    phrases emphasizing the main factors or causes behind events or trends

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.13
    0.7%
    393
    +0.12
    0.7%
    292
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    289
    +0.13
    0.02
    429
    +0.12
    0.02
    458
    +0.12
    0.01
    Negative Logits
    actly
    -1.59
    sizeof
    -1.55
    abe
    -1.50
    thirds
    -1.49
    ableness
    -1.48
    abilities
    -1.46
    abel
    -1.41
     many
    -1.40
     deepest
    -1.39
    Times
    -1.39
    POSITIVE LOGITS
     focused
    2.01
     because
    1.88
     concentrated
    1.79
     devoted
    1.76
     foc
    1.72
     geared
    1.71
     aimed
    1.69
     attributable
    1.62
     due
    1.62
     owing
    1.62
    Activations Density 0.253%

    No Known Activations