© 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. 10287
    Prev
    Next
    INDEX
    Explanations

    names of historical and biblical figures

    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
    348
    +0.30
    1.8%
    307
    +0.15
    0.9%
    410
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    71
    +0.30
    0.13
    156
    +0.15
    0.16
    410
    +0.15
    0.13
    Negative Logits
    ĥ½
    -2.83
    -2.81
     ↵
    -2.81
    ↵
    -2.81
    -2.81
    -2.81
    <|outofrange|>
    -2.81
    ↵
    -2.81
    ↵↵
    -2.81
    č↵
    -2.81
    POSITIVE LOGITS
    's
    2.01
     himself
    1.85
    ’
    1.80
    stadt
    1.76
     Lect
    1.70
     Brothers
    1.69
    enos
    1.67
     VIII
    1.67
    ston
    1.62
    ‘
    1.58
    Activations Density 1.759%

    No Known Activations