© 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. 1-RES-SM
    11. 25668
    Prev
    Next
    INDEX
    Explanations

    repeated use of the word "the" in various contexts

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    350
    +0.14
    0.8%
    23
    +0.14
    0.8%
    156
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    10
    +0.14
    0.04
    362
    +0.14
    0.04
    350
    +0.14
    0.03
    Negative Logits
    ŀ
    -3.11
    İ
    -3.08
    Ĥ
    -2.87
    ¢
    -2.80
    Ń
    -2.78
    ¶
    -2.76
    ģ
    -2.71
    ľĵ
    -2.69
    Ī
    -2.67
    ¾
    -2.67
    POSITIVE LOGITS
     entirety
    1.74
    utz
    1.72
    inkle
    1.71
     afternoon
    1.71
    apa
    1.66
    acket
    1.58
    illy
    1.57
     morning
    1.56
    amy
    1.43
    inth
    1.41
    Activations Density 0.455%

    No Known Activations