© 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. 2-RES-SM
    11. 13726
    Prev
    Next
    INDEX
    Explanations

    coding elements and structured data within text

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    237
    +0.13
    0.8%
    400
    +0.13
    0.8%
    161
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    237
    +0.13
    0.17
    337
    +0.13
    0.07
    240
    +0.13
    -0.06
    Negative Logits
    ·¸
    -1.87
    merce
    -1.50
    ake
    -1.50
    cester
    -1.49
    consin
    -1.43
    aches
    -1.32
    etine
    -1.32
    apolis
    -1.32
    mie
    -1.31
    uppose
    -1.31
    POSITIVE LOGITS
     dialogue
    1.31
    aligned
    1.31
    frequency
    1.30
     dialog
    1.29
    fileID
    1.28
    âĢ¦]
    1.28
    arguments
    1.27
    urg
    1.27
    consistent
    1.27
    fellow
    1.26
    Activations Density 7.712%

    No Known Activations